removed prompts files

This commit is contained in:
idittrich-valueon 2026-03-22 19:47:41 +01:00
parent 9a2e2b4f5f
commit 749e357f8e
349 changed files with 0 additions and 14111 deletions

View file

@ -1,119 +0,0 @@
# TASK: Plan Document Structure (Documents + Chapters)
This is a STRUCTURE PLANNING task. You define which documents to create and which chapters each document will have.
Chapter CONTENT will be generated in a later step - here you only plan the STRUCTURE and assign content references.
Return EXACTLY ONE complete JSON object. Do not generate multiple JSON objects, alternatives, or variations. Do not use separators like "---" between JSON objects.
## USER REQUEST (for context)
```
entwirf eine email mit dem inhalt des files davor
```
## AVAILABLE CONTENT PARTS
(No content parts available)
## CONTENT ASSIGNMENT RULE
CRITICAL: Every chapter MUST have contentParts assigned if it relates to documents/images/data from the user request.
If the user request mentions documents/images/data, then EVERY chapter that generates content related to those references MUST assign the relevant ContentParts explicitly.
Assignment logic:
- If chapter DISPLAYS a document/image → assign "object" format ContentPart with "caption"
- If chapter generates text content ABOUT a document/image/data → assign ContentPart with "instruction":
- Prefer "extracted" format if available (contains analyzed/extracted content)
- If only "object" format is available, use "object" format with "instruction" (to write ABOUT the image/document)
- If chapter's generationHint or purpose relates to a document/image/data mentioned in user request → it MUST have ContentParts assigned
- Multiple chapters might assign the same ContentPart (e.g., one chapter displays image, another writes about it)
- Use ContentPart IDs exactly as listed in AVAILABLE CONTENT PARTS above
- Empty contentParts are only allowed if chapter generates content WITHOUT referencing any documents/images/data from the user request
CRITICAL RULE: If the user request mentions BOTH:
a) Documents/images/data (listed in AVAILABLE CONTENT PARTS above), AND
b) Generic content types (article text, main content, body text, etc.)
Then chapters that generate those generic content types MUST assign the relevant ContentParts, because the content should relate to or be based on the provided documents/images/data.
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential information only
- AVOID verbose, lengthy, or repetitive text - be concise and direct
- Prioritize FACTS over filler text - no introductions like "In this chapter..."
- Minimize system resources: shorter content = faster processing
- Quality over quantity: precise, meaningful content rather than padding
## CHAPTER STRUCTURE REQUIREMENTS
- Generate chapters based on USER REQUEST - analyze what structure the user wants
- Create ONLY the minimum chapters needed to cover the user's request - avoid over-structuring
- HARD LIMIT: Maximum 7 chapters per document. If the topic can be covered in fewer, prefer fewer. Combine related topics into single chapters rather than creating many small ones.
- IMPORTANT: Each chapter MUST have ALL these fields:
- id: Unique identifier (e.g., "chapter_1")
- level: Heading level (1, 2, 3, etc.)
- title: Chapter title
- contentParts: Object mapping ContentPart IDs to usage instructions (MUST assign if chapter relates to documents/data from user request)
- generationHint: Description of what content to generate (including formatting/styling requirements)
- sections: Empty array [] (REQUIRED - sections are generated in next phase)
- contentParts: {"partId": {"instruction": "..."} or {"caption": "..."} or both} - Assign ContentParts as required by CONTENT ASSIGNMENT RULE above
- The "instruction" field for each ContentPart MUST contain ALL relevant details from the USER REQUEST that apply to content extraction for this specific chapter. Include all formatting rules, data requirements, constraints, and specifications mentioned in the user request that are relevant for processing this ContentPart in this chapter.
- generationHint: Keep CONCISE but include relevant details from the USER REQUEST. Focus on WHAT to generate, not HOW to phrase it verbosely.
- The number of chapters depends on the user request - create only what is requested. Do NOT create chapters for topics without available data.
CRITICAL: Only create chapters for CONTENT sections, not for formatting/styling requirements. Formatting/styling requirements to be included in each generationHint if needed.
## DOCUMENT STRUCTURE
For each document, determine:
- outputFormat: From USER REQUEST (explicit mention or infer from purpose/content type). Default: "txt". Multiple documents can have different formats.
- language: From USER REQUEST (map to ISO 639-1: de, en, fr, it...). Default: "de". Multiple documents can have different languages.
- chapters: Structure appropriately for the format (e.g., pptx=slides, docx=sections, xlsx=worksheets). Match format capabilities and constraints.
Required JSON fields:
- metadata: {"title": "...", "language": "..."}
- documents: Array with id, title, filename, outputFormat, language, chapters[]
- chapters: Array with id, level, title, contentParts, generationHint, sections[]
EXAMPLE STRUCTURE (for reference only - adapt to user request):
{
"metadata": {
"title": "Document Title",
"language": "de"
},
"documents": [{
"id": "doc_1",
"title": "Document Title",
"filename": "document.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Chapter Title",
"contentParts": {
"extracted_part_id": {
"instruction": "Use extracted content with ALL relevant details from user request"
}
},
"generationHint": "Detailed description including ALL relevant details from user request for this chapter",
"sections": []
}
]
}]
}
CRITICAL INSTRUCTIONS:
- Generate chapters based on USER REQUEST, NOT based on the example above
- The example shows the JSON structure format, NOT the required chapters
- Create only the chapters that match the user's request
- Adapt chapter titles and structure to match the user's specific request
- Determine outputFormat and language for each document by analyzing the USER REQUEST above
- The example shows placeholders "txt" and "de" - YOU MUST REPLACE THESE with actual values determined from the USER REQUEST
MANDATORY CONTENT ASSIGNMENT CHECK:
For each chapter, verify:
1. Does the user request mention documents/images/data? (e.g., "photo", "image", "document", "data", "based on", "about")
2. Does this chapter's generationHint, title, or purpose relate to those documents/images/data mentioned in step 1?
- Examples: "article about the photo", "text describing the image", "analysis of the document", "content based on the data"
- Even if chapter doesn't explicitly say "about the image", if user request mentions both the image AND this chapter's content type → relate them
3. If YES to both → chapter MUST have contentParts assigned (cannot be empty {})
4. If ContentPart is "object" format and chapter needs to write ABOUT it → assign with "instruction" field, not just "caption"
OUTPUT FORMAT: Start with { and end with }. Do NOT use markdown code fences (```json). Do NOT add explanatory text before or after the JSON. Return ONLY the JSON object itself.

View file

@ -1,41 +0,0 @@
{
"metadata": {
"title": "Email Entwurf",
"language": "de"
},
"documents": [
{
"id": "doc_1",
"title": "Email Entwurf",
"filename": "email_entwurf.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Einleitung",
"contentParts": {},
"generationHint": "Erstellen Sie eine kurze Einleitung für die E-Mail, die den Zweck der Nachricht erklärt.",
"sections": []
},
{
"id": "chapter_2",
"level": 1,
"title": "Hauptteil",
"contentParts": {},
"generationHint": "Formulieren Sie den Hauptinhalt der E-Mail basierend auf den Informationen, die im vorherigen Dokument enthalten waren.",
"sections": []
},
{
"id": "chapter_3",
"level": 1,
"title": "Schluss",
"contentParts": {},
"generationHint": "Schreiben Sie einen abschließenden Absatz, der die E-Mail höflich beendet und eventuell weitere Schritte oder Erwartungen erwähnt.",
"sections": []
}
]
}
]
}

View file

@ -1,41 +0,0 @@
{
"metadata": {
"title": "Email Entwurf",
"language": "de"
},
"documents": [
{
"id": "doc_1",
"title": "Email Entwurf",
"filename": "email_entwurf.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Einleitung",
"contentParts": {},
"generationHint": "Erstellen Sie eine kurze Einleitung für die E-Mail, die den Zweck der Nachricht erklärt.",
"sections": []
},
{
"id": "chapter_2",
"level": 1,
"title": "Hauptteil",
"contentParts": {},
"generationHint": "Formulieren Sie den Hauptinhalt der E-Mail basierend auf den Informationen, die im vorherigen Dokument enthalten waren.",
"sections": []
},
{
"id": "chapter_3",
"level": 1,
"title": "Schluss",
"contentParts": {},
"generationHint": "Schreiben Sie einen abschließenden Absatz, der die E-Mail höflich beendet und eventuell weitere Schritte oder Erwartungen erwähnt.",
"sections": []
}
]
}
]
}

View file

@ -1,119 +0,0 @@
# TASK: Plan Document Structure (Documents + Chapters)
This is a STRUCTURE PLANNING task. You define which documents to create and which chapters each document will have.
Chapter CONTENT will be generated in a later step - here you only plan the STRUCTURE and assign content references.
Return EXACTLY ONE complete JSON object. Do not generate multiple JSON objects, alternatives, or variations. Do not use separators like "---" between JSON objects.
## USER REQUEST (for context)
```
entwirf eine email mit dem inhalt des files davor
```
## AVAILABLE CONTENT PARTS
(No content parts available)
## CONTENT ASSIGNMENT RULE
CRITICAL: Every chapter MUST have contentParts assigned if it relates to documents/images/data from the user request.
If the user request mentions documents/images/data, then EVERY chapter that generates content related to those references MUST assign the relevant ContentParts explicitly.
Assignment logic:
- If chapter DISPLAYS a document/image → assign "object" format ContentPart with "caption"
- If chapter generates text content ABOUT a document/image/data → assign ContentPart with "instruction":
- Prefer "extracted" format if available (contains analyzed/extracted content)
- If only "object" format is available, use "object" format with "instruction" (to write ABOUT the image/document)
- If chapter's generationHint or purpose relates to a document/image/data mentioned in user request → it MUST have ContentParts assigned
- Multiple chapters might assign the same ContentPart (e.g., one chapter displays image, another writes about it)
- Use ContentPart IDs exactly as listed in AVAILABLE CONTENT PARTS above
- Empty contentParts are only allowed if chapter generates content WITHOUT referencing any documents/images/data from the user request
CRITICAL RULE: If the user request mentions BOTH:
a) Documents/images/data (listed in AVAILABLE CONTENT PARTS above), AND
b) Generic content types (article text, main content, body text, etc.)
Then chapters that generate those generic content types MUST assign the relevant ContentParts, because the content should relate to or be based on the provided documents/images/data.
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential information only
- AVOID verbose, lengthy, or repetitive text - be concise and direct
- Prioritize FACTS over filler text - no introductions like "In this chapter..."
- Minimize system resources: shorter content = faster processing
- Quality over quantity: precise, meaningful content rather than padding
## CHAPTER STRUCTURE REQUIREMENTS
- Generate chapters based on USER REQUEST - analyze what structure the user wants
- Create ONLY the minimum chapters needed to cover the user's request - avoid over-structuring
- HARD LIMIT: Maximum 7 chapters per document. If the topic can be covered in fewer, prefer fewer. Combine related topics into single chapters rather than creating many small ones.
- IMPORTANT: Each chapter MUST have ALL these fields:
- id: Unique identifier (e.g., "chapter_1")
- level: Heading level (1, 2, 3, etc.)
- title: Chapter title
- contentParts: Object mapping ContentPart IDs to usage instructions (MUST assign if chapter relates to documents/data from user request)
- generationHint: Description of what content to generate (including formatting/styling requirements)
- sections: Empty array [] (REQUIRED - sections are generated in next phase)
- contentParts: {"partId": {"instruction": "..."} or {"caption": "..."} or both} - Assign ContentParts as required by CONTENT ASSIGNMENT RULE above
- The "instruction" field for each ContentPart MUST contain ALL relevant details from the USER REQUEST that apply to content extraction for this specific chapter. Include all formatting rules, data requirements, constraints, and specifications mentioned in the user request that are relevant for processing this ContentPart in this chapter.
- generationHint: Keep CONCISE but include relevant details from the USER REQUEST. Focus on WHAT to generate, not HOW to phrase it verbosely.
- The number of chapters depends on the user request - create only what is requested. Do NOT create chapters for topics without available data.
CRITICAL: Only create chapters for CONTENT sections, not for formatting/styling requirements. Formatting/styling requirements to be included in each generationHint if needed.
## DOCUMENT STRUCTURE
For each document, determine:
- outputFormat: From USER REQUEST (explicit mention or infer from purpose/content type). Default: "txt". Multiple documents can have different formats.
- language: From USER REQUEST (map to ISO 639-1: de, en, fr, it...). Default: "de". Multiple documents can have different languages.
- chapters: Structure appropriately for the format (e.g., pptx=slides, docx=sections, xlsx=worksheets). Match format capabilities and constraints.
Required JSON fields:
- metadata: {"title": "...", "language": "..."}
- documents: Array with id, title, filename, outputFormat, language, chapters[]
- chapters: Array with id, level, title, contentParts, generationHint, sections[]
EXAMPLE STRUCTURE (for reference only - adapt to user request):
{
"metadata": {
"title": "Document Title",
"language": "de"
},
"documents": [{
"id": "doc_1",
"title": "Document Title",
"filename": "document.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Chapter Title",
"contentParts": {
"extracted_part_id": {
"instruction": "Use extracted content with ALL relevant details from user request"
}
},
"generationHint": "Detailed description including ALL relevant details from user request for this chapter",
"sections": []
}
]
}]
}
CRITICAL INSTRUCTIONS:
- Generate chapters based on USER REQUEST, NOT based on the example above
- The example shows the JSON structure format, NOT the required chapters
- Create only the chapters that match the user's request
- Adapt chapter titles and structure to match the user's specific request
- Determine outputFormat and language for each document by analyzing the USER REQUEST above
- The example shows placeholders "txt" and "de" - YOU MUST REPLACE THESE with actual values determined from the USER REQUEST
MANDATORY CONTENT ASSIGNMENT CHECK:
For each chapter, verify:
1. Does the user request mention documents/images/data? (e.g., "photo", "image", "document", "data", "based on", "about")
2. Does this chapter's generationHint, title, or purpose relate to those documents/images/data mentioned in step 1?
- Examples: "article about the photo", "text describing the image", "analysis of the document", "content based on the data"
- Even if chapter doesn't explicitly say "about the image", if user request mentions both the image AND this chapter's content type → relate them
3. If YES to both → chapter MUST have contentParts assigned (cannot be empty {})
4. If ContentPart is "object" format and chapter needs to write ABOUT it → assign with "instruction" field, not just "caption"
OUTPUT FORMAT: Start with { and end with }. Do NOT use markdown code fences (```json). Do NOT add explanatory text before or after the JSON. Return ONLY the JSON object itself.

View file

@ -1,41 +0,0 @@
{
"metadata": {
"title": "Email Entwurf",
"language": "de"
},
"documents": [
{
"id": "doc_1",
"title": "Email Entwurf",
"filename": "email_entwurf.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Einleitung",
"contentParts": {},
"generationHint": "Erstellen Sie eine Einleitung für die E-Mail, die den Zweck und den Kontext der Nachricht kurz beschreibt.",
"sections": []
},
{
"id": "chapter_2",
"level": 1,
"title": "Hauptinhalt",
"contentParts": {},
"generationHint": "Formulieren Sie den Hauptinhalt der E-Mail, der die wesentlichen Informationen oder Anfragen enthält.",
"sections": []
},
{
"id": "chapter_3",
"level": 1,
"title": "Abschluss",
"contentParts": {},
"generationHint": "Fügen Sie einen Abschluss hinzu, der die E-Mail höflich beendet und eventuell weitere Schritte oder Erwartungen erwähnt.",
"sections": []
}
]
}
]
}

View file

@ -1,41 +0,0 @@
{
"metadata": {
"title": "Email Entwurf",
"language": "de"
},
"documents": [
{
"id": "doc_1",
"title": "Email Entwurf",
"filename": "email_entwurf.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Einleitung",
"contentParts": {},
"generationHint": "Erstellen Sie eine Einleitung für die E-Mail, die den Zweck und den Kontext der Nachricht kurz beschreibt.",
"sections": []
},
{
"id": "chapter_2",
"level": 1,
"title": "Hauptinhalt",
"contentParts": {},
"generationHint": "Formulieren Sie den Hauptinhalt der E-Mail, der die wesentlichen Informationen oder Anfragen enthält.",
"sections": []
},
{
"id": "chapter_3",
"level": 1,
"title": "Abschluss",
"contentParts": {},
"generationHint": "Fügen Sie einen Abschluss hinzu, der die E-Mail höflich beendet und eventuell weitere Schritte oder Erwartungen erwähnt.",
"sections": []
}
]
}
]
}

View file

@ -1,119 +0,0 @@
# TASK: Plan Document Structure (Documents + Chapters)
This is a STRUCTURE PLANNING task. You define which documents to create and which chapters each document will have.
Chapter CONTENT will be generated in a later step - here you only plan the STRUCTURE and assign content references.
Return EXACTLY ONE complete JSON object. Do not generate multiple JSON objects, alternatives, or variations. Do not use separators like "---" between JSON objects.
## USER REQUEST (for context)
```
entwirf eine email mit dem inhalt des files davor
```
## AVAILABLE CONTENT PARTS
(No content parts available)
## CONTENT ASSIGNMENT RULE
CRITICAL: Every chapter MUST have contentParts assigned if it relates to documents/images/data from the user request.
If the user request mentions documents/images/data, then EVERY chapter that generates content related to those references MUST assign the relevant ContentParts explicitly.
Assignment logic:
- If chapter DISPLAYS a document/image → assign "object" format ContentPart with "caption"
- If chapter generates text content ABOUT a document/image/data → assign ContentPart with "instruction":
- Prefer "extracted" format if available (contains analyzed/extracted content)
- If only "object" format is available, use "object" format with "instruction" (to write ABOUT the image/document)
- If chapter's generationHint or purpose relates to a document/image/data mentioned in user request → it MUST have ContentParts assigned
- Multiple chapters might assign the same ContentPart (e.g., one chapter displays image, another writes about it)
- Use ContentPart IDs exactly as listed in AVAILABLE CONTENT PARTS above
- Empty contentParts are only allowed if chapter generates content WITHOUT referencing any documents/images/data from the user request
CRITICAL RULE: If the user request mentions BOTH:
a) Documents/images/data (listed in AVAILABLE CONTENT PARTS above), AND
b) Generic content types (article text, main content, body text, etc.)
Then chapters that generate those generic content types MUST assign the relevant ContentParts, because the content should relate to or be based on the provided documents/images/data.
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential information only
- AVOID verbose, lengthy, or repetitive text - be concise and direct
- Prioritize FACTS over filler text - no introductions like "In this chapter..."
- Minimize system resources: shorter content = faster processing
- Quality over quantity: precise, meaningful content rather than padding
## CHAPTER STRUCTURE REQUIREMENTS
- Generate chapters based on USER REQUEST - analyze what structure the user wants
- Create ONLY the minimum chapters needed to cover the user's request - avoid over-structuring
- HARD LIMIT: Maximum 7 chapters per document. If the topic can be covered in fewer, prefer fewer. Combine related topics into single chapters rather than creating many small ones.
- IMPORTANT: Each chapter MUST have ALL these fields:
- id: Unique identifier (e.g., "chapter_1")
- level: Heading level (1, 2, 3, etc.)
- title: Chapter title
- contentParts: Object mapping ContentPart IDs to usage instructions (MUST assign if chapter relates to documents/data from user request)
- generationHint: Description of what content to generate (including formatting/styling requirements)
- sections: Empty array [] (REQUIRED - sections are generated in next phase)
- contentParts: {"partId": {"instruction": "..."} or {"caption": "..."} or both} - Assign ContentParts as required by CONTENT ASSIGNMENT RULE above
- The "instruction" field for each ContentPart MUST contain ALL relevant details from the USER REQUEST that apply to content extraction for this specific chapter. Include all formatting rules, data requirements, constraints, and specifications mentioned in the user request that are relevant for processing this ContentPart in this chapter.
- generationHint: Keep CONCISE but include relevant details from the USER REQUEST. Focus on WHAT to generate, not HOW to phrase it verbosely.
- The number of chapters depends on the user request - create only what is requested. Do NOT create chapters for topics without available data.
CRITICAL: Only create chapters for CONTENT sections, not for formatting/styling requirements. Formatting/styling requirements to be included in each generationHint if needed.
## DOCUMENT STRUCTURE
For each document, determine:
- outputFormat: From USER REQUEST (explicit mention or infer from purpose/content type). Default: "txt". Multiple documents can have different formats.
- language: From USER REQUEST (map to ISO 639-1: de, en, fr, it...). Default: "de". Multiple documents can have different languages.
- chapters: Structure appropriately for the format (e.g., pptx=slides, docx=sections, xlsx=worksheets). Match format capabilities and constraints.
Required JSON fields:
- metadata: {"title": "...", "language": "..."}
- documents: Array with id, title, filename, outputFormat, language, chapters[]
- chapters: Array with id, level, title, contentParts, generationHint, sections[]
EXAMPLE STRUCTURE (for reference only - adapt to user request):
{
"metadata": {
"title": "Document Title",
"language": "de"
},
"documents": [{
"id": "doc_1",
"title": "Document Title",
"filename": "document.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Chapter Title",
"contentParts": {
"extracted_part_id": {
"instruction": "Use extracted content with ALL relevant details from user request"
}
},
"generationHint": "Detailed description including ALL relevant details from user request for this chapter",
"sections": []
}
]
}]
}
CRITICAL INSTRUCTIONS:
- Generate chapters based on USER REQUEST, NOT based on the example above
- The example shows the JSON structure format, NOT the required chapters
- Create only the chapters that match the user's request
- Adapt chapter titles and structure to match the user's specific request
- Determine outputFormat and language for each document by analyzing the USER REQUEST above
- The example shows placeholders "txt" and "de" - YOU MUST REPLACE THESE with actual values determined from the USER REQUEST
MANDATORY CONTENT ASSIGNMENT CHECK:
For each chapter, verify:
1. Does the user request mention documents/images/data? (e.g., "photo", "image", "document", "data", "based on", "about")
2. Does this chapter's generationHint, title, or purpose relate to those documents/images/data mentioned in step 1?
- Examples: "article about the photo", "text describing the image", "analysis of the document", "content based on the data"
- Even if chapter doesn't explicitly say "about the image", if user request mentions both the image AND this chapter's content type → relate them
3. If YES to both → chapter MUST have contentParts assigned (cannot be empty {})
4. If ContentPart is "object" format and chapter needs to write ABOUT it → assign with "instruction" field, not just "caption"
OUTPUT FORMAT: Start with { and end with }. Do NOT use markdown code fences (```json). Do NOT add explanatory text before or after the JSON. Return ONLY the JSON object itself.

View file

@ -1,41 +0,0 @@
{
"metadata": {
"title": "Email Entwurf",
"language": "de"
},
"documents": [
{
"id": "doc_1",
"title": "Email Entwurf",
"filename": "email_entwurf.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Einleitung",
"contentParts": {},
"generationHint": "Einleitung der E-Mail, Begrüßung und Einführung in das Thema.",
"sections": []
},
{
"id": "chapter_2",
"level": 1,
"title": "Hauptteil",
"contentParts": {},
"generationHint": "Hauptinhalt der E-Mail, Details und relevante Informationen.",
"sections": []
},
{
"id": "chapter_3",
"level": 1,
"title": "Schluss",
"contentParts": {},
"generationHint": "Abschluss der E-Mail, Zusammenfassung und Grußformel.",
"sections": []
}
]
}
]
}

View file

@ -1,41 +0,0 @@
{
"metadata": {
"title": "Email Entwurf",
"language": "de"
},
"documents": [
{
"id": "doc_1",
"title": "Email Entwurf",
"filename": "email_entwurf.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Einleitung",
"contentParts": {},
"generationHint": "Einleitung der E-Mail, Begrüßung und Einführung in das Thema.",
"sections": []
},
{
"id": "chapter_2",
"level": 1,
"title": "Hauptteil",
"contentParts": {},
"generationHint": "Hauptinhalt der E-Mail, Details und relevante Informationen.",
"sections": []
},
{
"id": "chapter_3",
"level": 1,
"title": "Schluss",
"contentParts": {},
"generationHint": "Abschluss der E-Mail, Zusammenfassung und Grußformel.",
"sections": []
}
]
}
]
}

View file

@ -1,119 +0,0 @@
# TASK: Plan Document Structure (Documents + Chapters)
This is a STRUCTURE PLANNING task. You define which documents to create and which chapters each document will have.
Chapter CONTENT will be generated in a later step - here you only plan the STRUCTURE and assign content references.
Return EXACTLY ONE complete JSON object. Do not generate multiple JSON objects, alternatives, or variations. Do not use separators like "---" between JSON objects.
## USER REQUEST (for context)
```
entwirf eine email mit dem inhalt des files davor
```
## AVAILABLE CONTENT PARTS
(No content parts available)
## CONTENT ASSIGNMENT RULE
CRITICAL: Every chapter MUST have contentParts assigned if it relates to documents/images/data from the user request.
If the user request mentions documents/images/data, then EVERY chapter that generates content related to those references MUST assign the relevant ContentParts explicitly.
Assignment logic:
- If chapter DISPLAYS a document/image → assign "object" format ContentPart with "caption"
- If chapter generates text content ABOUT a document/image/data → assign ContentPart with "instruction":
- Prefer "extracted" format if available (contains analyzed/extracted content)
- If only "object" format is available, use "object" format with "instruction" (to write ABOUT the image/document)
- If chapter's generationHint or purpose relates to a document/image/data mentioned in user request → it MUST have ContentParts assigned
- Multiple chapters might assign the same ContentPart (e.g., one chapter displays image, another writes about it)
- Use ContentPart IDs exactly as listed in AVAILABLE CONTENT PARTS above
- Empty contentParts are only allowed if chapter generates content WITHOUT referencing any documents/images/data from the user request
CRITICAL RULE: If the user request mentions BOTH:
a) Documents/images/data (listed in AVAILABLE CONTENT PARTS above), AND
b) Generic content types (article text, main content, body text, etc.)
Then chapters that generate those generic content types MUST assign the relevant ContentParts, because the content should relate to or be based on the provided documents/images/data.
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential information only
- AVOID verbose, lengthy, or repetitive text - be concise and direct
- Prioritize FACTS over filler text - no introductions like "In this chapter..."
- Minimize system resources: shorter content = faster processing
- Quality over quantity: precise, meaningful content rather than padding
## CHAPTER STRUCTURE REQUIREMENTS
- Generate chapters based on USER REQUEST - analyze what structure the user wants
- Create ONLY the minimum chapters needed to cover the user's request - avoid over-structuring
- HARD LIMIT: Maximum 7 chapters per document. If the topic can be covered in fewer, prefer fewer. Combine related topics into single chapters rather than creating many small ones.
- IMPORTANT: Each chapter MUST have ALL these fields:
- id: Unique identifier (e.g., "chapter_1")
- level: Heading level (1, 2, 3, etc.)
- title: Chapter title
- contentParts: Object mapping ContentPart IDs to usage instructions (MUST assign if chapter relates to documents/data from user request)
- generationHint: Description of what content to generate (including formatting/styling requirements)
- sections: Empty array [] (REQUIRED - sections are generated in next phase)
- contentParts: {"partId": {"instruction": "..."} or {"caption": "..."} or both} - Assign ContentParts as required by CONTENT ASSIGNMENT RULE above
- The "instruction" field for each ContentPart MUST contain ALL relevant details from the USER REQUEST that apply to content extraction for this specific chapter. Include all formatting rules, data requirements, constraints, and specifications mentioned in the user request that are relevant for processing this ContentPart in this chapter.
- generationHint: Keep CONCISE but include relevant details from the USER REQUEST. Focus on WHAT to generate, not HOW to phrase it verbosely.
- The number of chapters depends on the user request - create only what is requested. Do NOT create chapters for topics without available data.
CRITICAL: Only create chapters for CONTENT sections, not for formatting/styling requirements. Formatting/styling requirements to be included in each generationHint if needed.
## DOCUMENT STRUCTURE
For each document, determine:
- outputFormat: From USER REQUEST (explicit mention or infer from purpose/content type). Default: "txt". Multiple documents can have different formats.
- language: From USER REQUEST (map to ISO 639-1: de, en, fr, it...). Default: "de". Multiple documents can have different languages.
- chapters: Structure appropriately for the format (e.g., pptx=slides, docx=sections, xlsx=worksheets). Match format capabilities and constraints.
Required JSON fields:
- metadata: {"title": "...", "language": "..."}
- documents: Array with id, title, filename, outputFormat, language, chapters[]
- chapters: Array with id, level, title, contentParts, generationHint, sections[]
EXAMPLE STRUCTURE (for reference only - adapt to user request):
{
"metadata": {
"title": "Document Title",
"language": "de"
},
"documents": [{
"id": "doc_1",
"title": "Document Title",
"filename": "document.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Chapter Title",
"contentParts": {
"extracted_part_id": {
"instruction": "Use extracted content with ALL relevant details from user request"
}
},
"generationHint": "Detailed description including ALL relevant details from user request for this chapter",
"sections": []
}
]
}]
}
CRITICAL INSTRUCTIONS:
- Generate chapters based on USER REQUEST, NOT based on the example above
- The example shows the JSON structure format, NOT the required chapters
- Create only the chapters that match the user's request
- Adapt chapter titles and structure to match the user's specific request
- Determine outputFormat and language for each document by analyzing the USER REQUEST above
- The example shows placeholders "txt" and "de" - YOU MUST REPLACE THESE with actual values determined from the USER REQUEST
MANDATORY CONTENT ASSIGNMENT CHECK:
For each chapter, verify:
1. Does the user request mention documents/images/data? (e.g., "photo", "image", "document", "data", "based on", "about")
2. Does this chapter's generationHint, title, or purpose relate to those documents/images/data mentioned in step 1?
- Examples: "article about the photo", "text describing the image", "analysis of the document", "content based on the data"
- Even if chapter doesn't explicitly say "about the image", if user request mentions both the image AND this chapter's content type → relate them
3. If YES to both → chapter MUST have contentParts assigned (cannot be empty {})
4. If ContentPart is "object" format and chapter needs to write ABOUT it → assign with "instruction" field, not just "caption"
OUTPUT FORMAT: Start with { and end with }. Do NOT use markdown code fences (```json). Do NOT add explanatory text before or after the JSON. Return ONLY the JSON object itself.

View file

@ -1,41 +0,0 @@
{
"metadata": {
"title": "Email Entwurf",
"language": "de"
},
"documents": [
{
"id": "doc_1",
"title": "Email Entwurf",
"filename": "email_entwurf.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Einleitung",
"contentParts": {},
"generationHint": "Einleitung der E-Mail, die den Zweck und den Kontext der Nachricht kurz beschreibt.",
"sections": []
},
{
"id": "chapter_2",
"level": 1,
"title": "Hauptinhalt",
"contentParts": {},
"generationHint": "Hauptteil der E-Mail, der die wichtigsten Informationen und Details enthält.",
"sections": []
},
{
"id": "chapter_3",
"level": 1,
"title": "Abschluss",
"contentParts": {},
"generationHint": "Abschluss der E-Mail mit einer Zusammenfassung oder einem abschließenden Gedanken.",
"sections": []
}
]
}
]
}

View file

@ -1,41 +0,0 @@
{
"metadata": {
"title": "Email Entwurf",
"language": "de"
},
"documents": [
{
"id": "doc_1",
"title": "Email Entwurf",
"filename": "email_entwurf.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Einleitung",
"contentParts": {},
"generationHint": "Einleitung der E-Mail, die den Zweck und den Kontext der Nachricht kurz beschreibt.",
"sections": []
},
{
"id": "chapter_2",
"level": 1,
"title": "Hauptinhalt",
"contentParts": {},
"generationHint": "Hauptteil der E-Mail, der die wichtigsten Informationen und Details enthält.",
"sections": []
},
{
"id": "chapter_3",
"level": 1,
"title": "Abschluss",
"contentParts": {},
"generationHint": "Abschluss der E-Mail mit einer Zusammenfassung oder einem abschließenden Gedanken.",
"sections": []
}
]
}
]
}

View file

@ -1,73 +0,0 @@
TASK: Generate Chapter Sections Structure
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
CHAPTER: Einleitung (Level 1, ID: chapter_1)
GENERATION HINT: Einleitung der E-Mail, die den Zweck und den Kontext der Nachricht kurz beschreibt.
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT sections: Focus on essential information only
- AVOID creating too many sections - combine related content where possible
- Each section should serve a clear purpose with meaningful data
- If no relevant data exists for a topic, do NOT create a section for it
- Prefer ONE comprehensive section over multiple sparse sections
- HARD LIMIT: Maximum 5 sections per chapter. Combine related subtopics into single sections to stay within this limit.
**CRITICAL**: The chapter's generationHint above describes what content this chapter should generate. If the generationHint references documents/images/data, then EACH section that generates content for this chapter MUST assign the relevant ContentParts from AVAILABLE CONTENT PARTS below.
NOTE: Chapter already has a heading section. Do NOT generate a heading for the chapter title.
## SECTION INDEPENDENCE
- Each section is independent and self-contained
- One section does NOT have information about another section
- Each section must provide its own context and be understandable alone
AVAILABLE CONTENT PARTS:
(No content parts specified for this chapter)
## CONTENT ASSIGNMENT RULE - CRITICAL
If AVAILABLE CONTENT PARTS are listed above, then EVERY section that generates content related to those ContentParts MUST assign them explicitly.
**Assignment logic:**
- If section generates text content ABOUT a ContentPart → assign "extracted" format ContentPart with appropriate instruction
- If section DISPLAYS a ContentPart → assign "object" format ContentPart
- If section's generationHint or purpose relates to a ContentPart listed above → it MUST have contentPartIds assigned
- If chapter's generationHint references documents/images/data AND section generates content for that chapter → section MUST assign relevant ContentParts
- Empty contentPartIds [] are only allowed if section generates content WITHOUT referencing any available ContentParts AND WITHOUT relating to chapter's generationHint
## ACCEPTED CONTENT TYPES FOR THIS FORMAT
The document output format (txt) accepts only the following content types:
table, bullet_list, heading, paragraph, code_block
**CRITICAL**: Only create sections with content types from this list. Other types will fail.
useAiCall RULE (simple):
- useAiCall: true → Content needs AI processing (extract, transform, generate, filter, summarize)
- useAiCall: false → Content can be inserted directly without changes (Format is "object" or "reference")
RETURN JSON:
{
"sections": [
{
"id": "section_1",
"content_type": "table",
"contentPartIds": ["extracted_part_id"],
"generationHint": "Description of what to extract or generate",
"useAiCall": true,
"elements": []
}
]
}
**MANDATORY CONTENT ASSIGNMENT CHECK:**
For each section, verify:
1. Are ContentParts listed in AVAILABLE CONTENT PARTS above?
2. Does this section's generationHint or purpose relate to those ContentParts?
3. If YES to both → section MUST have contentPartIds assigned (cannot be empty [])
4. Assign ContentPart IDs exactly as listed in AVAILABLE CONTENT PARTS above
IMAGE SECTIONS:
- For image sections, always provide a "caption" field with a descriptive caption for the image.
Return only valid JSON. Do not include any explanatory text outside the JSON.

View file

@ -1,73 +0,0 @@
TASK: Generate Chapter Sections Structure
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
CHAPTER: Hauptinhalt (Level 1, ID: chapter_2)
GENERATION HINT: Hauptteil der E-Mail, der die wichtigsten Informationen und Details enthält.
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT sections: Focus on essential information only
- AVOID creating too many sections - combine related content where possible
- Each section should serve a clear purpose with meaningful data
- If no relevant data exists for a topic, do NOT create a section for it
- Prefer ONE comprehensive section over multiple sparse sections
- HARD LIMIT: Maximum 5 sections per chapter. Combine related subtopics into single sections to stay within this limit.
**CRITICAL**: The chapter's generationHint above describes what content this chapter should generate. If the generationHint references documents/images/data, then EACH section that generates content for this chapter MUST assign the relevant ContentParts from AVAILABLE CONTENT PARTS below.
NOTE: Chapter already has a heading section. Do NOT generate a heading for the chapter title.
## SECTION INDEPENDENCE
- Each section is independent and self-contained
- One section does NOT have information about another section
- Each section must provide its own context and be understandable alone
AVAILABLE CONTENT PARTS:
(No content parts specified for this chapter)
## CONTENT ASSIGNMENT RULE - CRITICAL
If AVAILABLE CONTENT PARTS are listed above, then EVERY section that generates content related to those ContentParts MUST assign them explicitly.
**Assignment logic:**
- If section generates text content ABOUT a ContentPart → assign "extracted" format ContentPart with appropriate instruction
- If section DISPLAYS a ContentPart → assign "object" format ContentPart
- If section's generationHint or purpose relates to a ContentPart listed above → it MUST have contentPartIds assigned
- If chapter's generationHint references documents/images/data AND section generates content for that chapter → section MUST assign relevant ContentParts
- Empty contentPartIds [] are only allowed if section generates content WITHOUT referencing any available ContentParts AND WITHOUT relating to chapter's generationHint
## ACCEPTED CONTENT TYPES FOR THIS FORMAT
The document output format (txt) accepts only the following content types:
table, bullet_list, heading, paragraph, code_block
**CRITICAL**: Only create sections with content types from this list. Other types will fail.
useAiCall RULE (simple):
- useAiCall: true → Content needs AI processing (extract, transform, generate, filter, summarize)
- useAiCall: false → Content can be inserted directly without changes (Format is "object" or "reference")
RETURN JSON:
{
"sections": [
{
"id": "section_1",
"content_type": "table",
"contentPartIds": ["extracted_part_id"],
"generationHint": "Description of what to extract or generate",
"useAiCall": true,
"elements": []
}
]
}
**MANDATORY CONTENT ASSIGNMENT CHECK:**
For each section, verify:
1. Are ContentParts listed in AVAILABLE CONTENT PARTS above?
2. Does this section's generationHint or purpose relate to those ContentParts?
3. If YES to both → section MUST have contentPartIds assigned (cannot be empty [])
4. Assign ContentPart IDs exactly as listed in AVAILABLE CONTENT PARTS above
IMAGE SECTIONS:
- For image sections, always provide a "caption" field with a descriptive caption for the image.
Return only valid JSON. Do not include any explanatory text outside the JSON.

View file

@ -1,73 +0,0 @@
TASK: Generate Chapter Sections Structure
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
CHAPTER: Abschluss (Level 1, ID: chapter_3)
GENERATION HINT: Abschluss der E-Mail mit einer Zusammenfassung oder einem abschließenden Gedanken.
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT sections: Focus on essential information only
- AVOID creating too many sections - combine related content where possible
- Each section should serve a clear purpose with meaningful data
- If no relevant data exists for a topic, do NOT create a section for it
- Prefer ONE comprehensive section over multiple sparse sections
- HARD LIMIT: Maximum 5 sections per chapter. Combine related subtopics into single sections to stay within this limit.
**CRITICAL**: The chapter's generationHint above describes what content this chapter should generate. If the generationHint references documents/images/data, then EACH section that generates content for this chapter MUST assign the relevant ContentParts from AVAILABLE CONTENT PARTS below.
NOTE: Chapter already has a heading section. Do NOT generate a heading for the chapter title.
## SECTION INDEPENDENCE
- Each section is independent and self-contained
- One section does NOT have information about another section
- Each section must provide its own context and be understandable alone
AVAILABLE CONTENT PARTS:
(No content parts specified for this chapter)
## CONTENT ASSIGNMENT RULE - CRITICAL
If AVAILABLE CONTENT PARTS are listed above, then EVERY section that generates content related to those ContentParts MUST assign them explicitly.
**Assignment logic:**
- If section generates text content ABOUT a ContentPart → assign "extracted" format ContentPart with appropriate instruction
- If section DISPLAYS a ContentPart → assign "object" format ContentPart
- If section's generationHint or purpose relates to a ContentPart listed above → it MUST have contentPartIds assigned
- If chapter's generationHint references documents/images/data AND section generates content for that chapter → section MUST assign relevant ContentParts
- Empty contentPartIds [] are only allowed if section generates content WITHOUT referencing any available ContentParts AND WITHOUT relating to chapter's generationHint
## ACCEPTED CONTENT TYPES FOR THIS FORMAT
The document output format (txt) accepts only the following content types:
table, bullet_list, heading, paragraph, code_block
**CRITICAL**: Only create sections with content types from this list. Other types will fail.
useAiCall RULE (simple):
- useAiCall: true → Content needs AI processing (extract, transform, generate, filter, summarize)
- useAiCall: false → Content can be inserted directly without changes (Format is "object" or "reference")
RETURN JSON:
{
"sections": [
{
"id": "section_1",
"content_type": "table",
"contentPartIds": ["extracted_part_id"],
"generationHint": "Description of what to extract or generate",
"useAiCall": true,
"elements": []
}
]
}
**MANDATORY CONTENT ASSIGNMENT CHECK:**
For each section, verify:
1. Are ContentParts listed in AVAILABLE CONTENT PARTS above?
2. Does this section's generationHint or purpose relate to those ContentParts?
3. If YES to both → section MUST have contentPartIds assigned (cannot be empty [])
4. Assign ContentPart IDs exactly as listed in AVAILABLE CONTENT PARTS above
IMAGE SECTIONS:
- For image sections, always provide a "caption" field with a descriptive caption for the image.
Return only valid JSON. Do not include any explanatory text outside the JSON.

View file

@ -1,16 +0,0 @@
```json
{
"sections": [
{
"id": "section_1",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Zusammenfassung der E-Mail-Inhalte und abschließende Gedanken",
"useAiCall": true,
"elements": [
"In der abschließenden Zusammenfassung der E-Mail möchten wir die wichtigsten Punkte noch einmal hervorheben. Es ist entscheidend, dass alle besprochenen Themen klar und prägnant zusammengefasst werden, um Missverständnisse zu vermeiden. Abschließend möchten wir betonen, wie wichtig die Zusammenarbeit und der kontinuierliche Dialog sind, um die gesteckten Ziele erfolgreich zu erreichen. Wir danken Ihnen für Ihre Aufmerksamkeit und freuen uns auf die weitere Zusammenarbeit."
]
}
]
}
```

View file

@ -1,39 +0,0 @@
```json
{
"sections": [
{
"id": "section_1",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Kurze Einführung in den Zweck der E-Mail",
"useAiCall": true,
"elements": [
"Diese E-Mail dient dazu, den Empfänger über den Zweck und den Kontext der Nachricht zu informieren. Sie bietet eine kurze Übersicht über die Hauptthemen, die in den folgenden Abschnitten behandelt werden."
]
},
{
"id": "section_2",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Kontext der Nachricht und warum sie wichtig ist",
"useAiCall": true,
"elements": [
"Der Kontext dieser Nachricht ist entscheidend, um die Relevanz und Dringlichkeit der behandelten Themen zu verstehen. Es wird erläutert, warum diese Informationen für den Empfänger von Bedeutung sind und welche Schritte möglicherweise erforderlich sind."
]
},
{
"id": "section_3",
"content_type": "bullet_list",
"contentPartIds": [],
"generationHint": "Hauptpunkte der Nachricht",
"useAiCall": true,
"elements": [
"Zweck der E-Mail",
"Wichtige Informationen und Updates",
"Erforderliche Aktionen oder Antworten",
"Relevante Fristen oder Termine"
]
}
]
}
```

View file

@ -1,66 +0,0 @@
```json
{
"sections": [
{
"id": "section_1",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Einleitung des Hauptinhalts mit einer kurzen Zusammenfassung der wichtigsten Punkte.",
"useAiCall": true,
"elements": [
"Der Hauptinhalt dieser E-Mail bietet eine umfassende Übersicht über die wesentlichen Informationen, die für den Empfänger von Bedeutung sind. In den folgenden Abschnitten werden die wichtigsten Details und Anweisungen klar und prägnant dargestellt."
]
},
{
"id": "section_2",
"content_type": "bullet_list",
"contentPartIds": [],
"generationHint": "Liste der wichtigsten Informationen oder Anweisungen.",
"useAiCall": true,
"elements": [
"Wichtiger Termin: 15. November 2023",
"Erforderliche Dokumente: Reisepass, Visum",
"Kontaktperson: Herr Müller, erreichbar unter mueller@example.com",
"Nächste Schritte: Anmeldung bis zum 10. November 2023"
]
},
{
"id": "section_3",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Detaillierte Beschreibung eines spezifischen Themas oder einer Anweisung.",
"useAiCall": true,
"elements": [
"Bitte beachten Sie, dass alle Teilnehmer vor dem 10. November 2023 registriert sein müssen, um an der Veranstaltung teilnehmen zu können. Die Registrierung erfolgt über das Online-Portal, das Ihnen per E-Mail zugesandt wurde. Stellen Sie sicher, dass alle erforderlichen Dokumente hochgeladen werden, um Verzögerungen zu vermeiden."
]
},
{
"id": "section_4",
"content_type": "table",
"contentPartIds": [],
"generationHint": "Tabelle mit den wichtigsten Terminen und Fristen.",
"useAiCall": true,
"elements": [
{
"headers": ["Ereignis", "Datum", "Details"],
"rows": [
["Anmeldeschluss", "10. November 2023", "Online-Registrierung erforderlich"],
["Veranstaltungstermin", "15. November 2023", "Beginn um 9:00 Uhr"],
["Dokumenteneinreichung", "Bis 12. November 2023", "Alle Dokumente müssen vollständig sein"]
]
}
]
},
{
"id": "section_5",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Abschluss des Hauptinhalts mit einem Aufruf zum Handeln oder einer abschließenden Bemerkung.",
"useAiCall": true,
"elements": [
"Wir freuen uns auf Ihre Teilnahme und stehen für Rückfragen jederzeit zur Verfügung. Bitte zögern Sie nicht, uns zu kontaktieren, falls Sie weitere Informationen benötigen. Ihre rechtzeitige Anmeldung wird sehr geschätzt."
]
}
]
}
```

View file

@ -1,56 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_1
- Content Type: paragraph
- Generation Hint: Kurze Einführung in den Zweck der E-Mail
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (paragraph).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "paragraph",
"content": {{"text": "This is paragraph text."}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Following sections:
- section_2 (paragraph):
- section_3 (bullet_list):

View file

@ -1,59 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_2
- Content Type: paragraph
- Generation Hint: Kontext der Nachricht und warum sie wichtig ist
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (paragraph).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "paragraph",
"content": {{"text": "This is paragraph text."}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Previous sections:
- section_1 (paragraph):
Following sections:
- section_3 (bullet_list):
- section_1 (paragraph):

View file

@ -1,60 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_3
- Content Type: bullet_list
- Generation Hint: Hauptpunkte der Nachricht
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (bullet_list).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "bullet_list",
"content": {{"items": ["Item 1", "Item 2", "Item 3"]}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Previous sections:
- section_1 (paragraph):
- section_2 (paragraph):
Following sections:
- section_1 (paragraph):
- section_2 (bullet_list):

View file

@ -1,56 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_1
- Content Type: paragraph
- Generation Hint: Einleitung des Hauptinhalts mit einer kurzen Zusammenfassung der wichtigsten Punkte.
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (paragraph).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "paragraph",
"content": {{"text": "This is paragraph text."}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Following sections:
- section_2 (paragraph):
- section_3 (bullet_list):

View file

@ -1,59 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_2
- Content Type: bullet_list
- Generation Hint: Liste der wichtigsten Informationen oder Anweisungen.
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (bullet_list).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "bullet_list",
"content": {{"items": ["Item 1", "Item 2", "Item 3"]}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Previous sections:
- section_1 (paragraph):
Following sections:
- section_3 (bullet_list):
- section_1 (paragraph):

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Diese E-Mail dient dazu, den Zweck und die wichtigsten Informationen des vorherigen Dokuments zusammenzufassen. Sie bietet eine kurze Übersicht über die Hauptpunkte und soll als Einführung in die detaillierteren Abschnitte des Dokuments dienen."
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Diese E-Mail dient dazu, den Zweck und die wichtigsten Informationen des vorherigen Dokuments zusammenzufassen. Sie bietet eine kurze Übersicht über die Hauptpunkte und soll als Einführung in die detaillierteren Abschnitte des Dokuments dienen."
}
}
]
}

View file

@ -1,60 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_3
- Content Type: paragraph
- Generation Hint: Detaillierte Beschreibung eines spezifischen Themas oder einer Anweisung.
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (paragraph).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "paragraph",
"content": {{"text": "This is paragraph text."}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Previous sections:
- section_1 (paragraph):
- section_2 (paragraph):
Following sections:
- section_1 (paragraph):
- section_2 (bullet_list):

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "In dieser Einleitung werden die Hauptpunkte des Dokuments zusammengefasst. Der Fokus liegt auf der klaren Darstellung der wichtigsten Informationen, um dem Leser einen schnellen Überblick zu verschaffen. Die folgenden Abschnitte bieten detaillierte Einblicke und ergänzende Details zu den hier genannten Themen."
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "In dieser Einleitung werden die Hauptpunkte des Dokuments zusammengefasst. Der Fokus liegt auf der klaren Darstellung der wichtigsten Informationen, um dem Leser einen schnellen Überblick zu verschaffen. Die folgenden Abschnitte bieten detaillierte Einblicke und ergänzende Details zu den hier genannten Themen."
}
}
]
}

View file

@ -1,60 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_4
- Content Type: table
- Generation Hint: Tabelle mit den wichtigsten Terminen und Fristen.
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (table).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "table",
"content": {{"headers": ["Column1", "Column2"], "rows": [["Value1", "Value2"], ["Value3", "Value4"]]}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Previous sections:
- section_2 (paragraph):
- section_3 (bullet_list):
Following sections:
- section_2 (bullet_list):
- section_3 (paragraph):

View file

@ -1,16 +0,0 @@
{
"elements": [
{
"type": "bullet_list",
"content": {
"items": [
"Betreff der E-Mail klar und prägnant formulieren.",
"Anrede personalisieren, um den Empfänger direkt anzusprechen.",
"Hauptinhalt der E-Mail in kurzen, prägnanten Sätzen darstellen.",
"Wichtige Informationen oder Anweisungen in Stichpunkten auflisten.",
"Abschließende Grußformel und Kontaktdaten angeben."
]
}
}
]
}

View file

@ -1,16 +0,0 @@
{
"elements": [
{
"type": "bullet_list",
"content": {
"items": [
"Betreff der E-Mail klar und prägnant formulieren.",
"Anrede personalisieren, um den Empfänger direkt anzusprechen.",
"Hauptinhalt der E-Mail in kurzen, prägnanten Sätzen darstellen.",
"Wichtige Informationen oder Anweisungen in Stichpunkten auflisten.",
"Abschließende Grußformel und Kontaktdaten angeben."
]
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Die Nachricht ist von entscheidender Bedeutung, da sie wichtige Informationen übermittelt, die für die Empfänger relevant sind. Sie dient dazu, Klarheit zu schaffen, Missverständnisse zu vermeiden und die Empfänger über wesentliche Entwicklungen oder Entscheidungen zu informieren. Eine präzise und gut formulierte Nachricht trägt dazu bei, dass alle Beteiligten auf dem gleichen Stand sind und entsprechend handeln können. Dies fördert die Effizienz und Effektivität in der Kommunikation und im weiteren Handeln."
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Die Nachricht ist von entscheidender Bedeutung, da sie wichtige Informationen übermittelt, die für die Empfänger relevant sind. Sie dient dazu, Klarheit zu schaffen, Missverständnisse zu vermeiden und die Empfänger über wesentliche Entwicklungen oder Entscheidungen zu informieren. Eine präzise und gut formulierte Nachricht trägt dazu bei, dass alle Beteiligten auf dem gleichen Stand sind und entsprechend handeln können. Dies fördert die Effizienz und Effektivität in der Kommunikation und im weiteren Handeln."
}
}
]
}

View file

@ -1,60 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_5
- Content Type: paragraph
- Generation Hint: Abschluss des Hauptinhalts mit einem Aufruf zum Handeln oder einer abschließenden Bemerkung.
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (paragraph).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "paragraph",
"content": {{"text": "This is paragraph text."}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Previous sections:
- section_3 (bullet_list):
- section_1 (paragraph):
Following sections:
- section_3 (paragraph):
- section_4 (table):

View file

@ -1,56 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_1
- Content Type: paragraph
- Generation Hint: Zusammenfassung der E-Mail-Inhalte und abschließende Gedanken
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (paragraph).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "paragraph",
"content": {{"text": "This is paragraph text."}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Following sections:
- section_2 (paragraph):
- section_3 (bullet_list):

View file

@ -1,16 +0,0 @@
{
"elements": [
{
"type": "bullet_list",
"content": {
"items": [
"Betreff der E-Mail klar und prägnant formulieren.",
"Einleitung mit einer kurzen Begrüßung und Vorstellung.",
"Hauptinhalt der Nachricht in klaren Punkten darstellen.",
"Wichtige Informationen oder Anfragen deutlich hervorheben.",
"Abschluss mit einem Dank und einer höflichen Verabschiedung."
]
}
}
]
}

View file

@ -1,16 +0,0 @@
{
"elements": [
{
"type": "bullet_list",
"content": {
"items": [
"Betreff der E-Mail klar und prägnant formulieren.",
"Einleitung mit einer kurzen Begrüßung und Vorstellung.",
"Hauptinhalt der Nachricht in klaren Punkten darstellen.",
"Wichtige Informationen oder Anfragen deutlich hervorheben.",
"Abschluss mit einem Dank und einer höflichen Verabschiedung."
]
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Vielen Dank, dass Sie sich die Zeit genommen haben, unseren Hauptinhalt zu lesen. Wir hoffen, dass die Informationen für Sie nützlich waren. Zögern Sie nicht, uns bei Fragen oder Anmerkungen zu kontaktieren. Wir freuen uns darauf, von Ihnen zu hören und stehen Ihnen jederzeit zur Verfügung, um Ihnen weiterzuhelfen."
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Vielen Dank, dass Sie sich die Zeit genommen haben, unseren Hauptinhalt zu lesen. Wir hoffen, dass die Informationen für Sie nützlich waren. Zögern Sie nicht, uns bei Fragen oder Anmerkungen zu kontaktieren. Wir freuen uns darauf, von Ihnen zu hören und stehen Ihnen jederzeit zur Verfügung, um Ihnen weiterzuhelfen."
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Beim Entwerfen einer E-Mail ist es wichtig, den Zweck klar zu definieren und die Nachricht prägnant zu halten. Beginnen Sie mit einer höflichen Anrede, gefolgt von einer kurzen Einführung, die den Kontext der E-Mail erklärt. Der Hauptteil sollte die wesentlichen Informationen oder Anfragen enthalten, klar strukturiert und leicht verständlich. Abschließend sollte die E-Mail mit einem freundlichen Gruß und, falls erforderlich, einem Hinweis auf mögliche nächste Schritte oder eine Antwortmöglichkeit enden. Achten Sie darauf, die E-Mail auf Rechtschreibung und Grammatik zu überprüfen, um Professionalität zu gewährleisten."
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Beim Entwerfen einer E-Mail ist es wichtig, den Zweck klar zu definieren und die Nachricht prägnant zu halten. Beginnen Sie mit einer höflichen Anrede, gefolgt von einer kurzen Einführung, die den Kontext der E-Mail erklärt. Der Hauptteil sollte die wesentlichen Informationen oder Anfragen enthalten, klar strukturiert und leicht verständlich. Abschließend sollte die E-Mail mit einem freundlichen Gruß und, falls erforderlich, einem Hinweis auf mögliche nächste Schritte oder eine Antwortmöglichkeit enden. Achten Sie darauf, die E-Mail auf Rechtschreibung und Grammatik zu überprüfen, um Professionalität zu gewährleisten."
}
}
]
}

View file

@ -1,17 +0,0 @@
{
"elements": [
{
"type": "table",
"content": {
"headers": ["Termin", "Frist"],
"rows": [
["Anmeldeschluss", "15. März 2024"],
["Zahlungsfrist", "30. März 2024"],
["Einreichung der Unterlagen", "10. April 2024"],
["Veranstaltungsbeginn", "20. April 2024"],
["Stornierungsfrist", "5. April 2024"]
]
}
}
]
}

View file

@ -1,35 +0,0 @@
{
"elements": [
{
"type": "table",
"content": {
"headers": [
"Termin",
"Frist"
],
"rows": [
[
"Anmeldeschluss",
"15. März 2024"
],
[
"Zahlungsfrist",
"30. März 2024"
],
[
"Einreichung der Unterlagen",
"10. April 2024"
],
[
"Veranstaltungsbeginn",
"20. April 2024"
],
[
"Stornierungsfrist",
"5. April 2024"
]
]
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Die E-Mail fasst die wichtigsten Punkte zusammen und bietet abschließende Gedanken. Sie enthält eine klare Übersicht über die besprochenen Themen und hebt die entscheidenden Informationen hervor. Abschließend wird ein Ausblick auf die nächsten Schritte gegeben, um die besprochenen Themen weiter zu verfolgen und die Kommunikation zu verbessern."
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Die E-Mail fasst die wichtigsten Punkte zusammen und bietet abschließende Gedanken. Sie enthält eine klare Übersicht über die besprochenen Themen und hebt die entscheidenden Informationen hervor. Abschließend wird ein Ausblick auf die nächsten Schritte gegeben, um die besprochenen Themen weiter zu verfolgen und die Kommunikation zu verbessern."
}
}
]
}

View file

@ -1,231 +0,0 @@
{
"metadata": {
"title": "Email Entwurf",
"language": "de"
},
"documents": [
{
"id": "doc_1",
"title": "Email Entwurf",
"filename": "email_entwurf.txt",
"outputFormat": "txt",
"language": "de",
"sections": [
{
"id": "chapter_1_heading",
"content_type": "heading",
"elements": [
{
"type": "heading",
"content": {
"text": "Einleitung",
"level": 1
}
}
]
},
{
"id": "section_1",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Kurze Einführung in den Zweck der E-Mail",
"useAiCall": true,
"elements": [
{
"type": "paragraph",
"content": {
"text": "Diese E-Mail dient dazu, den Zweck und die wichtigsten Informationen des vorherigen Dokuments zusammenzufassen. Sie bietet eine kurze Übersicht über die Hauptpunkte und soll als Einführung in die detaillierteren Abschnitte des Dokuments dienen."
}
}
]
},
{
"id": "section_2",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Kontext der Nachricht und warum sie wichtig ist",
"useAiCall": true,
"elements": [
{
"type": "paragraph",
"content": {
"text": "Die Nachricht ist von entscheidender Bedeutung, da sie wichtige Informationen übermittelt, die für die Empfänger relevant sind. Sie dient dazu, Klarheit zu schaffen, Missverständnisse zu vermeiden und die Empfänger über wesentliche Entwicklungen oder Entscheidungen zu informieren. Eine präzise und gut formulierte Nachricht trägt dazu bei, dass alle Beteiligten auf dem gleichen Stand sind und entsprechend handeln können. Dies fördert die Effizienz und Effektivität in der Kommunikation und im weiteren Handeln."
}
}
]
},
{
"id": "section_3",
"content_type": "bullet_list",
"contentPartIds": [],
"generationHint": "Hauptpunkte der Nachricht",
"useAiCall": true,
"elements": [
{
"type": "bullet_list",
"content": {
"items": [
"Betreff der E-Mail klar und prägnant formulieren.",
"Einleitung mit einer kurzen Begrüßung und Vorstellung.",
"Hauptinhalt der Nachricht in klaren Punkten darstellen.",
"Wichtige Informationen oder Anfragen deutlich hervorheben.",
"Abschluss mit einem Dank und einer höflichen Verabschiedung."
]
}
}
]
},
{
"id": "chapter_2_heading",
"content_type": "heading",
"elements": [
{
"type": "heading",
"content": {
"text": "Hauptinhalt",
"level": 1
}
}
]
},
{
"id": "section_1",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Einleitung des Hauptinhalts mit einer kurzen Zusammenfassung der wichtigsten Punkte.",
"useAiCall": true,
"elements": [
{
"type": "paragraph",
"content": {
"text": "In dieser Einleitung werden die Hauptpunkte des Dokuments zusammengefasst. Der Fokus liegt auf der klaren Darstellung der wichtigsten Informationen, um dem Leser einen schnellen Überblick zu verschaffen. Die folgenden Abschnitte bieten detaillierte Einblicke und ergänzende Details zu den hier genannten Themen."
}
}
]
},
{
"id": "section_2",
"content_type": "bullet_list",
"contentPartIds": [],
"generationHint": "Liste der wichtigsten Informationen oder Anweisungen.",
"useAiCall": true,
"elements": [
{
"type": "bullet_list",
"content": {
"items": [
"Betreff der E-Mail klar und prägnant formulieren.",
"Anrede personalisieren, um den Empfänger direkt anzusprechen.",
"Hauptinhalt der E-Mail in kurzen, prägnanten Sätzen darstellen.",
"Wichtige Informationen oder Anweisungen in Stichpunkten auflisten.",
"Abschließende Grußformel und Kontaktdaten angeben."
]
}
}
]
},
{
"id": "section_3",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Detaillierte Beschreibung eines spezifischen Themas oder einer Anweisung.",
"useAiCall": true,
"elements": [
{
"type": "paragraph",
"content": {
"text": "Beim Entwerfen einer E-Mail ist es wichtig, den Zweck klar zu definieren und die Nachricht prägnant zu halten. Beginnen Sie mit einer höflichen Anrede, gefolgt von einer kurzen Einführung, die den Kontext der E-Mail erklärt. Der Hauptteil sollte die wesentlichen Informationen oder Anfragen enthalten, klar strukturiert und leicht verständlich. Abschließend sollte die E-Mail mit einem freundlichen Gruß und, falls erforderlich, einem Hinweis auf mögliche nächste Schritte oder eine Antwortmöglichkeit enden. Achten Sie darauf, die E-Mail auf Rechtschreibung und Grammatik zu überprüfen, um Professionalität zu gewährleisten."
}
}
]
},
{
"id": "section_4",
"content_type": "table",
"contentPartIds": [],
"generationHint": "Tabelle mit den wichtigsten Terminen und Fristen.",
"useAiCall": true,
"elements": [
{
"type": "table",
"content": {
"headers": [
"Termin",
"Frist"
],
"rows": [
[
"Anmeldeschluss",
"15. März 2024"
],
[
"Zahlungsfrist",
"30. März 2024"
],
[
"Einreichung der Unterlagen",
"10. April 2024"
],
[
"Veranstaltungsbeginn",
"20. April 2024"
],
[
"Stornierungsfrist",
"5. April 2024"
]
]
}
}
]
},
{
"id": "section_5",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Abschluss des Hauptinhalts mit einem Aufruf zum Handeln oder einer abschließenden Bemerkung.",
"useAiCall": true,
"elements": [
{
"type": "paragraph",
"content": {
"text": "Vielen Dank, dass Sie sich die Zeit genommen haben, unseren Hauptinhalt zu lesen. Wir hoffen, dass die Informationen für Sie nützlich waren. Zögern Sie nicht, uns bei Fragen oder Anmerkungen zu kontaktieren. Wir freuen uns darauf, von Ihnen zu hören und stehen Ihnen jederzeit zur Verfügung, um Ihnen weiterzuhelfen."
}
}
]
},
{
"id": "chapter_3_heading",
"content_type": "heading",
"elements": [
{
"type": "heading",
"content": {
"text": "Abschluss",
"level": 1
}
}
]
},
{
"id": "section_1",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Zusammenfassung der E-Mail-Inhalte und abschließende Gedanken",
"useAiCall": true,
"elements": [
{
"type": "paragraph",
"content": {
"text": "Die E-Mail fasst die wichtigsten Punkte zusammen und bietet abschließende Gedanken. Sie enthält eine klare Übersicht über die besprochenen Themen und hebt die entscheidenden Informationen hervor. Abschließend wird ein Ausblick auf die nächsten Schritte gegeben, um die besprochenen Themen weiter zu verfolgen und die Kommunikation zu verbessern."
}
}
]
}
],
"metadata": {
"outputStyle": "document"
}
}
]
}

View file

@ -1,22 +0,0 @@
Compose an email based on this context:
-------
Compose an email. The upstream AI step did not produce output.
-------
Recipients: (not specified - this is a draft)
Style: business
Max length: 1000 characters
Available_Document_References:
- (no documents)
Based on the context, decide which documents to attach.
CRITICAL: Use EXACT document references from Available_Document_References above. For individual documents: ALWAYS use docItem:<documentId>:<filename> format (include filename)
Return JSON:
{
"subject": "subject line",
"body": "email body (HTML allowed)",
"attachments": ["docItem:<documentId>:<filename>"]
}

View file

@ -1,7 +0,0 @@
```json
{
"subject": "Follow-up on Upstream AI Process",
"body": "<p>Dear Team,</p><p>I hope this message finds you well. I am writing to inform you that the upstream AI step did not produce any output as expected. We need to address this issue promptly to ensure the continuity of our workflow.</p><p>Please let me know if there are any specific actions you would like me to take or if there are any updates on resolving this matter.</p><p>Thank you for your attention to this issue.</p><p>Best regards,<br>Your Name</p>",
"attachments": []
}
```

View file

@ -1,119 +0,0 @@
# TASK: Plan Document Structure (Documents + Chapters)
This is a STRUCTURE PLANNING task. You define which documents to create and which chapters each document will have.
Chapter CONTENT will be generated in a later step - here you only plan the STRUCTURE and assign content references.
Return EXACTLY ONE complete JSON object. Do not generate multiple JSON objects, alternatives, or variations. Do not use separators like "---" between JSON objects.
## USER REQUEST (for context)
```
entwirf eine email mit dem inhalt des files davor
```
## AVAILABLE CONTENT PARTS
(No content parts available)
## CONTENT ASSIGNMENT RULE
CRITICAL: Every chapter MUST have contentParts assigned if it relates to documents/images/data from the user request.
If the user request mentions documents/images/data, then EVERY chapter that generates content related to those references MUST assign the relevant ContentParts explicitly.
Assignment logic:
- If chapter DISPLAYS a document/image → assign "object" format ContentPart with "caption"
- If chapter generates text content ABOUT a document/image/data → assign ContentPart with "instruction":
- Prefer "extracted" format if available (contains analyzed/extracted content)
- If only "object" format is available, use "object" format with "instruction" (to write ABOUT the image/document)
- If chapter's generationHint or purpose relates to a document/image/data mentioned in user request → it MUST have ContentParts assigned
- Multiple chapters might assign the same ContentPart (e.g., one chapter displays image, another writes about it)
- Use ContentPart IDs exactly as listed in AVAILABLE CONTENT PARTS above
- Empty contentParts are only allowed if chapter generates content WITHOUT referencing any documents/images/data from the user request
CRITICAL RULE: If the user request mentions BOTH:
a) Documents/images/data (listed in AVAILABLE CONTENT PARTS above), AND
b) Generic content types (article text, main content, body text, etc.)
Then chapters that generate those generic content types MUST assign the relevant ContentParts, because the content should relate to or be based on the provided documents/images/data.
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential information only
- AVOID verbose, lengthy, or repetitive text - be concise and direct
- Prioritize FACTS over filler text - no introductions like "In this chapter..."
- Minimize system resources: shorter content = faster processing
- Quality over quantity: precise, meaningful content rather than padding
## CHAPTER STRUCTURE REQUIREMENTS
- Generate chapters based on USER REQUEST - analyze what structure the user wants
- Create ONLY the minimum chapters needed to cover the user's request - avoid over-structuring
- HARD LIMIT: Maximum 7 chapters per document. If the topic can be covered in fewer, prefer fewer. Combine related topics into single chapters rather than creating many small ones.
- IMPORTANT: Each chapter MUST have ALL these fields:
- id: Unique identifier (e.g., "chapter_1")
- level: Heading level (1, 2, 3, etc.)
- title: Chapter title
- contentParts: Object mapping ContentPart IDs to usage instructions (MUST assign if chapter relates to documents/data from user request)
- generationHint: Description of what content to generate (including formatting/styling requirements)
- sections: Empty array [] (REQUIRED - sections are generated in next phase)
- contentParts: {"partId": {"instruction": "..."} or {"caption": "..."} or both} - Assign ContentParts as required by CONTENT ASSIGNMENT RULE above
- The "instruction" field for each ContentPart MUST contain ALL relevant details from the USER REQUEST that apply to content extraction for this specific chapter. Include all formatting rules, data requirements, constraints, and specifications mentioned in the user request that are relevant for processing this ContentPart in this chapter.
- generationHint: Keep CONCISE but include relevant details from the USER REQUEST. Focus on WHAT to generate, not HOW to phrase it verbosely.
- The number of chapters depends on the user request - create only what is requested. Do NOT create chapters for topics without available data.
CRITICAL: Only create chapters for CONTENT sections, not for formatting/styling requirements. Formatting/styling requirements to be included in each generationHint if needed.
## DOCUMENT STRUCTURE
For each document, determine:
- outputFormat: From USER REQUEST (explicit mention or infer from purpose/content type). Default: "txt". Multiple documents can have different formats.
- language: From USER REQUEST (map to ISO 639-1: de, en, fr, it...). Default: "de". Multiple documents can have different languages.
- chapters: Structure appropriately for the format (e.g., pptx=slides, docx=sections, xlsx=worksheets). Match format capabilities and constraints.
Required JSON fields:
- metadata: {"title": "...", "language": "..."}
- documents: Array with id, title, filename, outputFormat, language, chapters[]
- chapters: Array with id, level, title, contentParts, generationHint, sections[]
EXAMPLE STRUCTURE (for reference only - adapt to user request):
{
"metadata": {
"title": "Document Title",
"language": "de"
},
"documents": [{
"id": "doc_1",
"title": "Document Title",
"filename": "document.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Chapter Title",
"contentParts": {
"extracted_part_id": {
"instruction": "Use extracted content with ALL relevant details from user request"
}
},
"generationHint": "Detailed description including ALL relevant details from user request for this chapter",
"sections": []
}
]
}]
}
CRITICAL INSTRUCTIONS:
- Generate chapters based on USER REQUEST, NOT based on the example above
- The example shows the JSON structure format, NOT the required chapters
- Create only the chapters that match the user's request
- Adapt chapter titles and structure to match the user's specific request
- Determine outputFormat and language for each document by analyzing the USER REQUEST above
- The example shows placeholders "txt" and "de" - YOU MUST REPLACE THESE with actual values determined from the USER REQUEST
MANDATORY CONTENT ASSIGNMENT CHECK:
For each chapter, verify:
1. Does the user request mention documents/images/data? (e.g., "photo", "image", "document", "data", "based on", "about")
2. Does this chapter's generationHint, title, or purpose relate to those documents/images/data mentioned in step 1?
- Examples: "article about the photo", "text describing the image", "analysis of the document", "content based on the data"
- Even if chapter doesn't explicitly say "about the image", if user request mentions both the image AND this chapter's content type → relate them
3. If YES to both → chapter MUST have contentParts assigned (cannot be empty {})
4. If ContentPart is "object" format and chapter needs to write ABOUT it → assign with "instruction" field, not just "caption"
OUTPUT FORMAT: Start with { and end with }. Do NOT use markdown code fences (```json). Do NOT add explanatory text before or after the JSON. Return ONLY the JSON object itself.

View file

@ -1,41 +0,0 @@
{
"metadata": {
"title": "Email Entwurf",
"language": "de"
},
"documents": [
{
"id": "doc_1",
"title": "Email Entwurf",
"filename": "email_entwurf.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Einleitung",
"contentParts": {},
"generationHint": "Einleitung der E-Mail, die den Zweck und den Kontext der Nachricht kurz beschreibt.",
"sections": []
},
{
"id": "chapter_2",
"level": 1,
"title": "Hauptinhalt",
"contentParts": {},
"generationHint": "Hauptteil der E-Mail, der die wesentlichen Informationen und Details enthält, die kommuniziert werden sollen.",
"sections": []
},
{
"id": "chapter_3",
"level": 1,
"title": "Schlussfolgerung",
"contentParts": {},
"generationHint": "Abschluss der E-Mail mit einer Zusammenfassung oder einem abschließenden Gedanken.",
"sections": []
}
]
}
]
}

View file

@ -1,41 +0,0 @@
{
"metadata": {
"title": "Email Entwurf",
"language": "de"
},
"documents": [
{
"id": "doc_1",
"title": "Email Entwurf",
"filename": "email_entwurf.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Einleitung",
"contentParts": {},
"generationHint": "Einleitung der E-Mail, die den Zweck und den Kontext der Nachricht kurz beschreibt.",
"sections": []
},
{
"id": "chapter_2",
"level": 1,
"title": "Hauptinhalt",
"contentParts": {},
"generationHint": "Hauptteil der E-Mail, der die wesentlichen Informationen und Details enthält, die kommuniziert werden sollen.",
"sections": []
},
{
"id": "chapter_3",
"level": 1,
"title": "Schlussfolgerung",
"contentParts": {},
"generationHint": "Abschluss der E-Mail mit einer Zusammenfassung oder einem abschließenden Gedanken.",
"sections": []
}
]
}
]
}

View file

@ -1,73 +0,0 @@
TASK: Generate Chapter Sections Structure
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
CHAPTER: Einleitung (Level 1, ID: chapter_1)
GENERATION HINT: Einleitung der E-Mail, die den Zweck und den Kontext der Nachricht kurz beschreibt.
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT sections: Focus on essential information only
- AVOID creating too many sections - combine related content where possible
- Each section should serve a clear purpose with meaningful data
- If no relevant data exists for a topic, do NOT create a section for it
- Prefer ONE comprehensive section over multiple sparse sections
- HARD LIMIT: Maximum 5 sections per chapter. Combine related subtopics into single sections to stay within this limit.
**CRITICAL**: The chapter's generationHint above describes what content this chapter should generate. If the generationHint references documents/images/data, then EACH section that generates content for this chapter MUST assign the relevant ContentParts from AVAILABLE CONTENT PARTS below.
NOTE: Chapter already has a heading section. Do NOT generate a heading for the chapter title.
## SECTION INDEPENDENCE
- Each section is independent and self-contained
- One section does NOT have information about another section
- Each section must provide its own context and be understandable alone
AVAILABLE CONTENT PARTS:
(No content parts specified for this chapter)
## CONTENT ASSIGNMENT RULE - CRITICAL
If AVAILABLE CONTENT PARTS are listed above, then EVERY section that generates content related to those ContentParts MUST assign them explicitly.
**Assignment logic:**
- If section generates text content ABOUT a ContentPart → assign "extracted" format ContentPart with appropriate instruction
- If section DISPLAYS a ContentPart → assign "object" format ContentPart
- If section's generationHint or purpose relates to a ContentPart listed above → it MUST have contentPartIds assigned
- If chapter's generationHint references documents/images/data AND section generates content for that chapter → section MUST assign relevant ContentParts
- Empty contentPartIds [] are only allowed if section generates content WITHOUT referencing any available ContentParts AND WITHOUT relating to chapter's generationHint
## ACCEPTED CONTENT TYPES FOR THIS FORMAT
The document output format (txt) accepts only the following content types:
table, bullet_list, heading, paragraph, code_block
**CRITICAL**: Only create sections with content types from this list. Other types will fail.
useAiCall RULE (simple):
- useAiCall: true → Content needs AI processing (extract, transform, generate, filter, summarize)
- useAiCall: false → Content can be inserted directly without changes (Format is "object" or "reference")
RETURN JSON:
{
"sections": [
{
"id": "section_1",
"content_type": "table",
"contentPartIds": ["extracted_part_id"],
"generationHint": "Description of what to extract or generate",
"useAiCall": true,
"elements": []
}
]
}
**MANDATORY CONTENT ASSIGNMENT CHECK:**
For each section, verify:
1. Are ContentParts listed in AVAILABLE CONTENT PARTS above?
2. Does this section's generationHint or purpose relate to those ContentParts?
3. If YES to both → section MUST have contentPartIds assigned (cannot be empty [])
4. Assign ContentPart IDs exactly as listed in AVAILABLE CONTENT PARTS above
IMAGE SECTIONS:
- For image sections, always provide a "caption" field with a descriptive caption for the image.
Return only valid JSON. Do not include any explanatory text outside the JSON.

View file

@ -1,73 +0,0 @@
TASK: Generate Chapter Sections Structure
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
CHAPTER: Hauptinhalt (Level 1, ID: chapter_2)
GENERATION HINT: Hauptteil der E-Mail, der die wesentlichen Informationen und Details enthält, die kommuniziert werden sollen.
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT sections: Focus on essential information only
- AVOID creating too many sections - combine related content where possible
- Each section should serve a clear purpose with meaningful data
- If no relevant data exists for a topic, do NOT create a section for it
- Prefer ONE comprehensive section over multiple sparse sections
- HARD LIMIT: Maximum 5 sections per chapter. Combine related subtopics into single sections to stay within this limit.
**CRITICAL**: The chapter's generationHint above describes what content this chapter should generate. If the generationHint references documents/images/data, then EACH section that generates content for this chapter MUST assign the relevant ContentParts from AVAILABLE CONTENT PARTS below.
NOTE: Chapter already has a heading section. Do NOT generate a heading for the chapter title.
## SECTION INDEPENDENCE
- Each section is independent and self-contained
- One section does NOT have information about another section
- Each section must provide its own context and be understandable alone
AVAILABLE CONTENT PARTS:
(No content parts specified for this chapter)
## CONTENT ASSIGNMENT RULE - CRITICAL
If AVAILABLE CONTENT PARTS are listed above, then EVERY section that generates content related to those ContentParts MUST assign them explicitly.
**Assignment logic:**
- If section generates text content ABOUT a ContentPart → assign "extracted" format ContentPart with appropriate instruction
- If section DISPLAYS a ContentPart → assign "object" format ContentPart
- If section's generationHint or purpose relates to a ContentPart listed above → it MUST have contentPartIds assigned
- If chapter's generationHint references documents/images/data AND section generates content for that chapter → section MUST assign relevant ContentParts
- Empty contentPartIds [] are only allowed if section generates content WITHOUT referencing any available ContentParts AND WITHOUT relating to chapter's generationHint
## ACCEPTED CONTENT TYPES FOR THIS FORMAT
The document output format (txt) accepts only the following content types:
table, bullet_list, heading, paragraph, code_block
**CRITICAL**: Only create sections with content types from this list. Other types will fail.
useAiCall RULE (simple):
- useAiCall: true → Content needs AI processing (extract, transform, generate, filter, summarize)
- useAiCall: false → Content can be inserted directly without changes (Format is "object" or "reference")
RETURN JSON:
{
"sections": [
{
"id": "section_1",
"content_type": "table",
"contentPartIds": ["extracted_part_id"],
"generationHint": "Description of what to extract or generate",
"useAiCall": true,
"elements": []
}
]
}
**MANDATORY CONTENT ASSIGNMENT CHECK:**
For each section, verify:
1. Are ContentParts listed in AVAILABLE CONTENT PARTS above?
2. Does this section's generationHint or purpose relate to those ContentParts?
3. If YES to both → section MUST have contentPartIds assigned (cannot be empty [])
4. Assign ContentPart IDs exactly as listed in AVAILABLE CONTENT PARTS above
IMAGE SECTIONS:
- For image sections, always provide a "caption" field with a descriptive caption for the image.
Return only valid JSON. Do not include any explanatory text outside the JSON.

View file

@ -1,73 +0,0 @@
TASK: Generate Chapter Sections Structure
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
CHAPTER: Schlussfolgerung (Level 1, ID: chapter_3)
GENERATION HINT: Abschluss der E-Mail mit einer Zusammenfassung oder einem abschließenden Gedanken.
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT sections: Focus on essential information only
- AVOID creating too many sections - combine related content where possible
- Each section should serve a clear purpose with meaningful data
- If no relevant data exists for a topic, do NOT create a section for it
- Prefer ONE comprehensive section over multiple sparse sections
- HARD LIMIT: Maximum 5 sections per chapter. Combine related subtopics into single sections to stay within this limit.
**CRITICAL**: The chapter's generationHint above describes what content this chapter should generate. If the generationHint references documents/images/data, then EACH section that generates content for this chapter MUST assign the relevant ContentParts from AVAILABLE CONTENT PARTS below.
NOTE: Chapter already has a heading section. Do NOT generate a heading for the chapter title.
## SECTION INDEPENDENCE
- Each section is independent and self-contained
- One section does NOT have information about another section
- Each section must provide its own context and be understandable alone
AVAILABLE CONTENT PARTS:
(No content parts specified for this chapter)
## CONTENT ASSIGNMENT RULE - CRITICAL
If AVAILABLE CONTENT PARTS are listed above, then EVERY section that generates content related to those ContentParts MUST assign them explicitly.
**Assignment logic:**
- If section generates text content ABOUT a ContentPart → assign "extracted" format ContentPart with appropriate instruction
- If section DISPLAYS a ContentPart → assign "object" format ContentPart
- If section's generationHint or purpose relates to a ContentPart listed above → it MUST have contentPartIds assigned
- If chapter's generationHint references documents/images/data AND section generates content for that chapter → section MUST assign relevant ContentParts
- Empty contentPartIds [] are only allowed if section generates content WITHOUT referencing any available ContentParts AND WITHOUT relating to chapter's generationHint
## ACCEPTED CONTENT TYPES FOR THIS FORMAT
The document output format (txt) accepts only the following content types:
table, bullet_list, heading, paragraph, code_block
**CRITICAL**: Only create sections with content types from this list. Other types will fail.
useAiCall RULE (simple):
- useAiCall: true → Content needs AI processing (extract, transform, generate, filter, summarize)
- useAiCall: false → Content can be inserted directly without changes (Format is "object" or "reference")
RETURN JSON:
{
"sections": [
{
"id": "section_1",
"content_type": "table",
"contentPartIds": ["extracted_part_id"],
"generationHint": "Description of what to extract or generate",
"useAiCall": true,
"elements": []
}
]
}
**MANDATORY CONTENT ASSIGNMENT CHECK:**
For each section, verify:
1. Are ContentParts listed in AVAILABLE CONTENT PARTS above?
2. Does this section's generationHint or purpose relate to those ContentParts?
3. If YES to both → section MUST have contentPartIds assigned (cannot be empty [])
4. Assign ContentPart IDs exactly as listed in AVAILABLE CONTENT PARTS above
IMAGE SECTIONS:
- For image sections, always provide a "caption" field with a descriptive caption for the image.
Return only valid JSON. Do not include any explanatory text outside the JSON.

View file

@ -1,19 +0,0 @@
```json
{
"sections": [
{
"id": "section_1",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Abschluss der E-Mail mit einer Zusammenfassung oder einem abschließenden Gedanken",
"useAiCall": true,
"elements": [
{
"type": "text",
"content": "In der Schlussfolgerung fassen wir die wichtigsten Punkte zusammen, die in der E-Mail behandelt wurden. Es ist entscheidend, den Empfänger mit einem klaren und prägnanten Abschluss zu hinterlassen, der die Hauptbotschaft verstärkt und eventuell nächste Schritte oder Erwartungen klärt. Ein abschließender Gedanke könnte auch eine Einladung zur weiteren Diskussion oder Zusammenarbeit sein, um die Beziehung zu stärken und zukünftige Kommunikation zu fördern."
}
]
}
]
}
```

View file

@ -1,38 +0,0 @@
```json
{
"sections": [
{
"id": "section_1",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Einleitung zur E-Mail, die den Zweck und den Kontext der Nachricht beschreibt",
"useAiCall": true,
"elements": [
"Diese E-Mail dient dazu, den Empfänger über die neuesten Entwicklungen in unserem Projekt zu informieren. Wir möchten Ihnen einen Überblick über die aktuellen Fortschritte geben und die nächsten Schritte erläutern. Ihr Feedback ist uns wichtig, um sicherzustellen, dass wir auf dem richtigen Weg sind."
]
},
{
"id": "section_2",
"content_type": "bullet_list",
"contentPartIds": [],
"generationHint": "Hauptziele der E-Mail zusammenfassen",
"useAiCall": true,
"elements": [
"Information über den aktuellen Projektstatus",
"Darstellung der nächsten Schritte",
"Einladung zur Rückmeldung und Diskussion"
]
},
{
"id": "section_3",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Kontext der Nachricht im Rahmen des größeren Projekts",
"useAiCall": true,
"elements": [
"Das Projekt, an dem wir arbeiten, ist ein wesentlicher Bestandteil unserer Unternehmensstrategie für das kommende Jahr. Es zielt darauf ab, unsere Marktposition zu stärken und neue Geschäftsmöglichkeiten zu erschließen."
]
}
]
}
```

View file

@ -1,87 +0,0 @@
```json
{
"sections": [
{
"id": "section_1",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Einleitung und Überblick über die wesentlichen Informationen der E-Mail",
"useAiCall": true,
"elements": [
{
"type": "text",
"content": "In diesem Abschnitt wird der Hauptinhalt der E-Mail vorgestellt. Hier werden die wichtigsten Informationen und Details zusammengefasst, die der Empfänger wissen muss. Der Fokus liegt darauf, die Kernaussagen klar und prägnant zu vermitteln."
}
]
},
{
"id": "section_2",
"content_type": "bullet_list",
"contentPartIds": [],
"generationHint": "Auflistung der Hauptpunkte oder Themen, die in der E-Mail behandelt werden",
"useAiCall": true,
"elements": [
{
"type": "item",
"content": "Ziel der E-Mail"
},
{
"type": "item",
"content": "Wichtige Termine oder Fristen"
},
{
"type": "item",
"content": "Erforderliche Maßnahmen oder Aktionen"
}
]
},
{
"id": "section_3",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Detaillierte Informationen zu einem spezifischen Thema oder Punkt",
"useAiCall": true,
"elements": [
{
"type": "text",
"content": "Dieser Abschnitt bietet eine detaillierte Erklärung zu einem bestimmten Thema, das in der E-Mail behandelt wird. Hier werden alle relevanten Details und Hintergrundinformationen bereitgestellt, um ein umfassendes Verständnis zu gewährleisten."
}
]
},
{
"id": "section_4",
"content_type": "table",
"contentPartIds": [],
"generationHint": "Tabellarische Darstellung von Daten oder Informationen, die in der E-Mail enthalten sind",
"useAiCall": true,
"elements": [
{
"type": "row",
"content": ["Thema", "Beschreibung"]
},
{
"type": "row",
"content": ["Projektstatus", "Aktueller Stand des Projekts und nächste Schritte"]
},
{
"type": "row",
"content": ["Budget", "Übersicht über die aktuellen Ausgaben und geplante Budgets"]
}
]
},
{
"id": "section_5",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Zusammenfassung und abschließende Bemerkungen",
"useAiCall": true,
"elements": [
{
"type": "text",
"content": "Zum Abschluss der E-Mail werden die wichtigsten Punkte noch einmal zusammengefasst. Es wird darauf hingewiesen, welche nächsten Schritte erforderlich sind und welche Fristen eingehalten werden müssen."
}
]
}
]
}
```

View file

@ -1,56 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_1
- Content Type: paragraph
- Generation Hint: Einleitung zur E-Mail, die den Zweck und den Kontext der Nachricht beschreibt
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (paragraph).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "paragraph",
"content": {{"text": "This is paragraph text."}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Following sections:
- section_2 (bullet_list):
- section_3 (paragraph):

View file

@ -1,59 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_2
- Content Type: bullet_list
- Generation Hint: Hauptziele der E-Mail zusammenfassen
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (bullet_list).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "bullet_list",
"content": {{"items": ["Item 1", "Item 2", "Item 3"]}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Previous sections:
- section_1 (paragraph):
Following sections:
- section_3 (paragraph):
- section_1 (paragraph):

View file

@ -1,60 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_3
- Content Type: paragraph
- Generation Hint: Kontext der Nachricht im Rahmen des größeren Projekts
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (paragraph).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "paragraph",
"content": {{"text": "This is paragraph text."}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Previous sections:
- section_1 (paragraph):
- section_2 (bullet_list):
Following sections:
- section_1 (paragraph):
- section_2 (bullet_list):

View file

@ -1,56 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_1
- Content Type: paragraph
- Generation Hint: Einleitung und Überblick über die wesentlichen Informationen der E-Mail
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (paragraph).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "paragraph",
"content": {{"text": "This is paragraph text."}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Following sections:
- section_2 (bullet_list):
- section_3 (paragraph):

View file

@ -1,59 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_2
- Content Type: bullet_list
- Generation Hint: Auflistung der Hauptpunkte oder Themen, die in der E-Mail behandelt werden
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (bullet_list).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "bullet_list",
"content": {{"items": ["Item 1", "Item 2", "Item 3"]}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Previous sections:
- section_1 (paragraph):
Following sections:
- section_3 (paragraph):
- section_1 (paragraph):

View file

@ -1,14 +0,0 @@
{
"elements": [
{
"type": "bullet_list",
"content": {
"items": [
"Wichtige Informationen klar und prägnant kommunizieren",
"Empfänger über spezifische Aktionen oder Entscheidungen informieren",
"Erwartungen oder nächste Schritte deutlich machen"
]
}
}
]
}

View file

@ -1,14 +0,0 @@
{
"elements": [
{
"type": "bullet_list",
"content": {
"items": [
"Wichtige Informationen klar und prägnant kommunizieren",
"Empfänger über spezifische Aktionen oder Entscheidungen informieren",
"Erwartungen oder nächste Schritte deutlich machen"
]
}
}
]
}

View file

@ -1,60 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_3
- Content Type: paragraph
- Generation Hint: Detaillierte Informationen zu einem spezifischen Thema oder Punkt
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (paragraph).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "paragraph",
"content": {{"text": "This is paragraph text."}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Previous sections:
- section_1 (paragraph):
- section_2 (bullet_list):
Following sections:
- section_1 (paragraph):
- section_2 (bullet_list):

View file

@ -1,16 +0,0 @@
{
"elements": [
{
"type": "bullet_list",
"content": {
"items": [
"Begrüßung und Vorstellung des Absenders",
"Zweck der E-Mail und Hauptthemen",
"Wichtige Informationen oder Updates",
"Anfragen oder benötigte Aktionen vom Empfänger",
"Abschluss und Dankesworte"
]
}
}
]
}

View file

@ -1,16 +0,0 @@
{
"elements": [
{
"type": "bullet_list",
"content": {
"items": [
"Begrüßung und Vorstellung des Absenders",
"Zweck der E-Mail und Hauptthemen",
"Wichtige Informationen oder Updates",
"Anfragen oder benötigte Aktionen vom Empfänger",
"Abschluss und Dankesworte"
]
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Diese E-Mail dient dazu, den Zweck und den Kontext der Nachricht klar darzustellen. Sie enthält wichtige Informationen, die für das Verständnis des Themas notwendig sind. Bitte lesen Sie die folgenden Abschnitte sorgfältig, um alle relevanten Details zu erfassen."
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Diese E-Mail dient dazu, den Zweck und den Kontext der Nachricht klar darzustellen. Sie enthält wichtige Informationen, die für das Verständnis des Themas notwendig sind. Bitte lesen Sie die folgenden Abschnitte sorgfältig, um alle relevanten Details zu erfassen."
}
}
]
}

View file

@ -1,60 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_4
- Content Type: table
- Generation Hint: Tabellarische Darstellung von Daten oder Informationen, die in der E-Mail enthalten sind
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (table).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "table",
"content": {{"headers": ["Column1", "Column2"], "rows": [["Value1", "Value2"], ["Value3", "Value4"]]}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Previous sections:
- section_2 (bullet_list):
- section_3 (paragraph):
Following sections:
- section_2 (bullet_list):
- section_3 (paragraph):

View file

@ -1,60 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_5
- Content Type: paragraph
- Generation Hint: Zusammenfassung und abschließende Bemerkungen
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (paragraph).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "paragraph",
"content": {{"text": "This is paragraph text."}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Previous sections:
- section_3 (paragraph):
- section_1 (paragraph):
Following sections:
- section_3 (paragraph):
- section_4 (table):

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Diese E-Mail bietet einen Überblick über die wichtigsten Informationen und Themen, die im vorherigen Dokument behandelt wurden. Sie fasst die wesentlichen Punkte zusammen, um den Empfängern einen schnellen und effektiven Einblick zu ermöglichen."
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Diese E-Mail bietet einen Überblick über die wichtigsten Informationen und Themen, die im vorherigen Dokument behandelt wurden. Sie fasst die wesentlichen Punkte zusammen, um den Empfängern einen schnellen und effektiven Einblick zu ermöglichen."
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Im Rahmen des Projekts ist es entscheidend, die Kommunikation klar und präzise zu halten. Die Nachricht sollte den aktuellen Fortschritt, die nächsten Schritte und eventuelle Herausforderungen adressieren. Es ist wichtig, alle relevanten Stakeholder auf dem Laufenden zu halten und sicherzustellen, dass alle notwendigen Ressourcen und Informationen bereitgestellt werden, um die Projektziele effizient zu erreichen."
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Im Rahmen des Projekts ist es entscheidend, die Kommunikation klar und präzise zu halten. Die Nachricht sollte den aktuellen Fortschritt, die nächsten Schritte und eventuelle Herausforderungen adressieren. Es ist wichtig, alle relevanten Stakeholder auf dem Laufenden zu halten und sicherzustellen, dass alle notwendigen Ressourcen und Informationen bereitgestellt werden, um die Projektziele effizient zu erreichen."
}
}
]
}

View file

@ -1,56 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_1
- Content Type: paragraph
- Generation Hint: Abschluss der E-Mail mit einer Zusammenfassung oder einem abschließenden Gedanken
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (paragraph).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "paragraph",
"content": {{"text": "This is paragraph text."}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Following sections:
- section_2 (bullet_list):
- section_3 (paragraph):

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Abschließend möchte ich mich für Ihre Aufmerksamkeit und Unterstützung bedanken. Ich freue mich auf Ihre Rückmeldung und hoffe auf eine positive Zusammenarbeit. Bei weiteren Fragen stehe ich Ihnen gerne zur Verfügung."
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Abschließend möchte ich mich für Ihre Aufmerksamkeit und Unterstützung bedanken. Ich freue mich auf Ihre Rückmeldung und hoffe auf eine positive Zusammenarbeit. Bei weiteren Fragen stehe ich Ihnen gerne zur Verfügung."
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Beim Entwerfen einer E-Mail ist es wichtig, den Zweck klar zu definieren und die Nachricht präzise zu formulieren. Beginnen Sie mit einer höflichen Anrede, gefolgt von einer kurzen Einführung in das Thema. Der Hauptteil sollte die wesentlichen Informationen enthalten, die der Empfänger benötigt, um die Nachricht zu verstehen und darauf zu reagieren. Abschließend sollte die E-Mail mit einem klaren Aufruf zum Handeln oder einer Zusammenfassung der nächsten Schritte enden. Höfliche Grüße und eine Signatur runden die E-Mail ab."
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Beim Entwerfen einer E-Mail ist es wichtig, den Zweck klar zu definieren und die Nachricht präzise zu formulieren. Beginnen Sie mit einer höflichen Anrede, gefolgt von einer kurzen Einführung in das Thema. Der Hauptteil sollte die wesentlichen Informationen enthalten, die der Empfänger benötigt, um die Nachricht zu verstehen und darauf zu reagieren. Abschließend sollte die E-Mail mit einem klaren Aufruf zum Handeln oder einer Zusammenfassung der nächsten Schritte enden. Höfliche Grüße und eine Signatur runden die E-Mail ab."
}
}
]
}

View file

@ -1,17 +0,0 @@
{
"elements": [
{
"type": "table",
"content": {
"headers": ["Betreff", "Inhalt"],
"rows": [
["Anrede", "Sehr geehrte Damen und Herren,"],
["Einleitung", "Ich hoffe, diese Nachricht trifft Sie wohl."],
["Hauptteil", "Im Anhang finden Sie die erforderlichen Informationen."],
["Schluss", "Vielen Dank für Ihre Aufmerksamkeit."],
["Grüße", "Mit freundlichen Grüßen, [Ihr Name]"]
]
}
}
]
}

View file

@ -1,35 +0,0 @@
{
"elements": [
{
"type": "table",
"content": {
"headers": [
"Betreff",
"Inhalt"
],
"rows": [
[
"Anrede",
"Sehr geehrte Damen und Herren,"
],
[
"Einleitung",
"Ich hoffe, diese Nachricht trifft Sie wohl."
],
[
"Hauptteil",
"Im Anhang finden Sie die erforderlichen Informationen."
],
[
"Schluss",
"Vielen Dank für Ihre Aufmerksamkeit."
],
[
"Grüße",
"Mit freundlichen Grüßen, [Ihr Name]"
]
]
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Zusammenfassend lässt sich sagen, dass die besprochenen Themen eine klare Richtung für zukünftige Maßnahmen vorgeben. Die Analyse der Daten zeigt deutliche Trends, die berücksichtigt werden müssen, um effektive Strategien zu entwickeln. Abschließend ist es wichtig, die gewonnenen Erkenntnisse in konkrete Handlungen umzusetzen, um die gewünschten Ergebnisse zu erzielen. Eine kontinuierliche Überprüfung und Anpassung der Strategien wird empfohlen, um auf Veränderungen im Umfeld flexibel reagieren zu können."
}
}
]
}

View file

@ -1,10 +0,0 @@
{
"elements": [
{
"type": "paragraph",
"content": {
"text": "Zusammenfassend lässt sich sagen, dass die besprochenen Themen eine klare Richtung für zukünftige Maßnahmen vorgeben. Die Analyse der Daten zeigt deutliche Trends, die berücksichtigt werden müssen, um effektive Strategien zu entwickeln. Abschließend ist es wichtig, die gewonnenen Erkenntnisse in konkrete Handlungen umzusetzen, um die gewünschten Ergebnisse zu erzielen. Eine kontinuierliche Überprüfung und Anpassung der Strategien wird empfohlen, um auf Veränderungen im Umfeld flexibel reagieren zu können."
}
}
]
}

View file

@ -1,229 +0,0 @@
{
"metadata": {
"title": "Email Entwurf",
"language": "de"
},
"documents": [
{
"id": "doc_1",
"title": "Email Entwurf",
"filename": "email_entwurf.txt",
"outputFormat": "txt",
"language": "de",
"sections": [
{
"id": "chapter_1_heading",
"content_type": "heading",
"elements": [
{
"type": "heading",
"content": {
"text": "Einleitung",
"level": 1
}
}
]
},
{
"id": "section_1",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Einleitung zur E-Mail, die den Zweck und den Kontext der Nachricht beschreibt",
"useAiCall": true,
"elements": [
{
"type": "paragraph",
"content": {
"text": "Diese E-Mail dient dazu, den Zweck und den Kontext der Nachricht klar darzustellen. Sie enthält wichtige Informationen, die für das Verständnis des Themas notwendig sind. Bitte lesen Sie die folgenden Abschnitte sorgfältig, um alle relevanten Details zu erfassen."
}
}
]
},
{
"id": "section_2",
"content_type": "bullet_list",
"contentPartIds": [],
"generationHint": "Hauptziele der E-Mail zusammenfassen",
"useAiCall": true,
"elements": [
{
"type": "bullet_list",
"content": {
"items": [
"Wichtige Informationen klar und prägnant kommunizieren",
"Empfänger über spezifische Aktionen oder Entscheidungen informieren",
"Erwartungen oder nächste Schritte deutlich machen"
]
}
}
]
},
{
"id": "section_3",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Kontext der Nachricht im Rahmen des größeren Projekts",
"useAiCall": true,
"elements": [
{
"type": "paragraph",
"content": {
"text": "Im Rahmen des Projekts ist es entscheidend, die Kommunikation klar und präzise zu halten. Die Nachricht sollte den aktuellen Fortschritt, die nächsten Schritte und eventuelle Herausforderungen adressieren. Es ist wichtig, alle relevanten Stakeholder auf dem Laufenden zu halten und sicherzustellen, dass alle notwendigen Ressourcen und Informationen bereitgestellt werden, um die Projektziele effizient zu erreichen."
}
}
]
},
{
"id": "chapter_2_heading",
"content_type": "heading",
"elements": [
{
"type": "heading",
"content": {
"text": "Hauptinhalt",
"level": 1
}
}
]
},
{
"id": "section_1",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Einleitung und Überblick über die wesentlichen Informationen der E-Mail",
"useAiCall": true,
"elements": [
{
"type": "paragraph",
"content": {
"text": "Diese E-Mail bietet einen Überblick über die wichtigsten Informationen und Themen, die im vorherigen Dokument behandelt wurden. Sie fasst die wesentlichen Punkte zusammen, um den Empfängern einen schnellen und effektiven Einblick zu ermöglichen."
}
}
]
},
{
"id": "section_2",
"content_type": "bullet_list",
"contentPartIds": [],
"generationHint": "Auflistung der Hauptpunkte oder Themen, die in der E-Mail behandelt werden",
"useAiCall": true,
"elements": [
{
"type": "bullet_list",
"content": {
"items": [
"Begrüßung und Vorstellung des Absenders",
"Zweck der E-Mail und Hauptthemen",
"Wichtige Informationen oder Updates",
"Anfragen oder benötigte Aktionen vom Empfänger",
"Abschluss und Dankesworte"
]
}
}
]
},
{
"id": "section_3",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Detaillierte Informationen zu einem spezifischen Thema oder Punkt",
"useAiCall": true,
"elements": [
{
"type": "paragraph",
"content": {
"text": "Beim Entwerfen einer E-Mail ist es wichtig, den Zweck klar zu definieren und die Nachricht präzise zu formulieren. Beginnen Sie mit einer höflichen Anrede, gefolgt von einer kurzen Einführung in das Thema. Der Hauptteil sollte die wesentlichen Informationen enthalten, die der Empfänger benötigt, um die Nachricht zu verstehen und darauf zu reagieren. Abschließend sollte die E-Mail mit einem klaren Aufruf zum Handeln oder einer Zusammenfassung der nächsten Schritte enden. Höfliche Grüße und eine Signatur runden die E-Mail ab."
}
}
]
},
{
"id": "section_4",
"content_type": "table",
"contentPartIds": [],
"generationHint": "Tabellarische Darstellung von Daten oder Informationen, die in der E-Mail enthalten sind",
"useAiCall": true,
"elements": [
{
"type": "table",
"content": {
"headers": [
"Betreff",
"Inhalt"
],
"rows": [
[
"Anrede",
"Sehr geehrte Damen und Herren,"
],
[
"Einleitung",
"Ich hoffe, diese Nachricht trifft Sie wohl."
],
[
"Hauptteil",
"Im Anhang finden Sie die erforderlichen Informationen."
],
[
"Schluss",
"Vielen Dank für Ihre Aufmerksamkeit."
],
[
"Grüße",
"Mit freundlichen Grüßen, [Ihr Name]"
]
]
}
}
]
},
{
"id": "section_5",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Zusammenfassung und abschließende Bemerkungen",
"useAiCall": true,
"elements": [
{
"type": "paragraph",
"content": {
"text": "Zusammenfassend lässt sich sagen, dass die besprochenen Themen eine klare Richtung für zukünftige Maßnahmen vorgeben. Die Analyse der Daten zeigt deutliche Trends, die berücksichtigt werden müssen, um effektive Strategien zu entwickeln. Abschließend ist es wichtig, die gewonnenen Erkenntnisse in konkrete Handlungen umzusetzen, um die gewünschten Ergebnisse zu erzielen. Eine kontinuierliche Überprüfung und Anpassung der Strategien wird empfohlen, um auf Veränderungen im Umfeld flexibel reagieren zu können."
}
}
]
},
{
"id": "chapter_3_heading",
"content_type": "heading",
"elements": [
{
"type": "heading",
"content": {
"text": "Schlussfolgerung",
"level": 1
}
}
]
},
{
"id": "section_1",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Abschluss der E-Mail mit einer Zusammenfassung oder einem abschließenden Gedanken",
"useAiCall": true,
"elements": [
{
"type": "paragraph",
"content": {
"text": "Abschließend möchte ich mich für Ihre Aufmerksamkeit und Unterstützung bedanken. Ich freue mich auf Ihre Rückmeldung und hoffe auf eine positive Zusammenarbeit. Bei weiteren Fragen stehe ich Ihnen gerne zur Verfügung."
}
}
]
}
],
"metadata": {
"outputStyle": "document"
}
}
]
}

View file

@ -1,119 +0,0 @@
# TASK: Plan Document Structure (Documents + Chapters)
This is a STRUCTURE PLANNING task. You define which documents to create and which chapters each document will have.
Chapter CONTENT will be generated in a later step - here you only plan the STRUCTURE and assign content references.
Return EXACTLY ONE complete JSON object. Do not generate multiple JSON objects, alternatives, or variations. Do not use separators like "---" between JSON objects.
## USER REQUEST (for context)
```
entwirf eine email mit dem inhalt des files davor
```
## AVAILABLE CONTENT PARTS
(No content parts available)
## CONTENT ASSIGNMENT RULE
CRITICAL: Every chapter MUST have contentParts assigned if it relates to documents/images/data from the user request.
If the user request mentions documents/images/data, then EVERY chapter that generates content related to those references MUST assign the relevant ContentParts explicitly.
Assignment logic:
- If chapter DISPLAYS a document/image → assign "object" format ContentPart with "caption"
- If chapter generates text content ABOUT a document/image/data → assign ContentPart with "instruction":
- Prefer "extracted" format if available (contains analyzed/extracted content)
- If only "object" format is available, use "object" format with "instruction" (to write ABOUT the image/document)
- If chapter's generationHint or purpose relates to a document/image/data mentioned in user request → it MUST have ContentParts assigned
- Multiple chapters might assign the same ContentPart (e.g., one chapter displays image, another writes about it)
- Use ContentPart IDs exactly as listed in AVAILABLE CONTENT PARTS above
- Empty contentParts are only allowed if chapter generates content WITHOUT referencing any documents/images/data from the user request
CRITICAL RULE: If the user request mentions BOTH:
a) Documents/images/data (listed in AVAILABLE CONTENT PARTS above), AND
b) Generic content types (article text, main content, body text, etc.)
Then chapters that generate those generic content types MUST assign the relevant ContentParts, because the content should relate to or be based on the provided documents/images/data.
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential information only
- AVOID verbose, lengthy, or repetitive text - be concise and direct
- Prioritize FACTS over filler text - no introductions like "In this chapter..."
- Minimize system resources: shorter content = faster processing
- Quality over quantity: precise, meaningful content rather than padding
## CHAPTER STRUCTURE REQUIREMENTS
- Generate chapters based on USER REQUEST - analyze what structure the user wants
- Create ONLY the minimum chapters needed to cover the user's request - avoid over-structuring
- HARD LIMIT: Maximum 7 chapters per document. If the topic can be covered in fewer, prefer fewer. Combine related topics into single chapters rather than creating many small ones.
- IMPORTANT: Each chapter MUST have ALL these fields:
- id: Unique identifier (e.g., "chapter_1")
- level: Heading level (1, 2, 3, etc.)
- title: Chapter title
- contentParts: Object mapping ContentPart IDs to usage instructions (MUST assign if chapter relates to documents/data from user request)
- generationHint: Description of what content to generate (including formatting/styling requirements)
- sections: Empty array [] (REQUIRED - sections are generated in next phase)
- contentParts: {"partId": {"instruction": "..."} or {"caption": "..."} or both} - Assign ContentParts as required by CONTENT ASSIGNMENT RULE above
- The "instruction" field for each ContentPart MUST contain ALL relevant details from the USER REQUEST that apply to content extraction for this specific chapter. Include all formatting rules, data requirements, constraints, and specifications mentioned in the user request that are relevant for processing this ContentPart in this chapter.
- generationHint: Keep CONCISE but include relevant details from the USER REQUEST. Focus on WHAT to generate, not HOW to phrase it verbosely.
- The number of chapters depends on the user request - create only what is requested. Do NOT create chapters for topics without available data.
CRITICAL: Only create chapters for CONTENT sections, not for formatting/styling requirements. Formatting/styling requirements to be included in each generationHint if needed.
## DOCUMENT STRUCTURE
For each document, determine:
- outputFormat: From USER REQUEST (explicit mention or infer from purpose/content type). Default: "txt". Multiple documents can have different formats.
- language: From USER REQUEST (map to ISO 639-1: de, en, fr, it...). Default: "de". Multiple documents can have different languages.
- chapters: Structure appropriately for the format (e.g., pptx=slides, docx=sections, xlsx=worksheets). Match format capabilities and constraints.
Required JSON fields:
- metadata: {"title": "...", "language": "..."}
- documents: Array with id, title, filename, outputFormat, language, chapters[]
- chapters: Array with id, level, title, contentParts, generationHint, sections[]
EXAMPLE STRUCTURE (for reference only - adapt to user request):
{
"metadata": {
"title": "Document Title",
"language": "de"
},
"documents": [{
"id": "doc_1",
"title": "Document Title",
"filename": "document.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Chapter Title",
"contentParts": {
"extracted_part_id": {
"instruction": "Use extracted content with ALL relevant details from user request"
}
},
"generationHint": "Detailed description including ALL relevant details from user request for this chapter",
"sections": []
}
]
}]
}
CRITICAL INSTRUCTIONS:
- Generate chapters based on USER REQUEST, NOT based on the example above
- The example shows the JSON structure format, NOT the required chapters
- Create only the chapters that match the user's request
- Adapt chapter titles and structure to match the user's specific request
- Determine outputFormat and language for each document by analyzing the USER REQUEST above
- The example shows placeholders "txt" and "de" - YOU MUST REPLACE THESE with actual values determined from the USER REQUEST
MANDATORY CONTENT ASSIGNMENT CHECK:
For each chapter, verify:
1. Does the user request mention documents/images/data? (e.g., "photo", "image", "document", "data", "based on", "about")
2. Does this chapter's generationHint, title, or purpose relate to those documents/images/data mentioned in step 1?
- Examples: "article about the photo", "text describing the image", "analysis of the document", "content based on the data"
- Even if chapter doesn't explicitly say "about the image", if user request mentions both the image AND this chapter's content type → relate them
3. If YES to both → chapter MUST have contentParts assigned (cannot be empty {})
4. If ContentPart is "object" format and chapter needs to write ABOUT it → assign with "instruction" field, not just "caption"
OUTPUT FORMAT: Start with { and end with }. Do NOT use markdown code fences (```json). Do NOT add explanatory text before or after the JSON. Return ONLY the JSON object itself.

View file

@ -1,41 +0,0 @@
{
"metadata": {
"title": "Email Entwurf",
"language": "de"
},
"documents": [
{
"id": "doc_1",
"title": "Email Entwurf",
"filename": "email_entwurf.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Einleitung",
"contentParts": {},
"generationHint": "Einleitung der E-Mail mit einer Begrüßung und Vorstellung des Themas.",
"sections": []
},
{
"id": "chapter_2",
"level": 1,
"title": "Hauptteil",
"contentParts": {},
"generationHint": "Hauptinhalt der E-Mail, der die wesentlichen Informationen und Anliegen beschreibt.",
"sections": []
},
{
"id": "chapter_3",
"level": 1,
"title": "Schluss",
"contentParts": {},
"generationHint": "Abschluss der E-Mail mit einer Zusammenfassung und einem freundlichen Gruß.",
"sections": []
}
]
}
]
}

View file

@ -1,41 +0,0 @@
{
"metadata": {
"title": "Email Entwurf",
"language": "de"
},
"documents": [
{
"id": "doc_1",
"title": "Email Entwurf",
"filename": "email_entwurf.txt",
"outputFormat": "txt",
"language": "de",
"chapters": [
{
"id": "chapter_1",
"level": 1,
"title": "Einleitung",
"contentParts": {},
"generationHint": "Einleitung der E-Mail mit einer Begrüßung und Vorstellung des Themas.",
"sections": []
},
{
"id": "chapter_2",
"level": 1,
"title": "Hauptteil",
"contentParts": {},
"generationHint": "Hauptinhalt der E-Mail, der die wesentlichen Informationen und Anliegen beschreibt.",
"sections": []
},
{
"id": "chapter_3",
"level": 1,
"title": "Schluss",
"contentParts": {},
"generationHint": "Abschluss der E-Mail mit einer Zusammenfassung und einem freundlichen Gruß.",
"sections": []
}
]
}
]
}

View file

@ -1,73 +0,0 @@
TASK: Generate Chapter Sections Structure
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
CHAPTER: Einleitung (Level 1, ID: chapter_1)
GENERATION HINT: Einleitung der E-Mail mit einer Begrüßung und Vorstellung des Themas.
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT sections: Focus on essential information only
- AVOID creating too many sections - combine related content where possible
- Each section should serve a clear purpose with meaningful data
- If no relevant data exists for a topic, do NOT create a section for it
- Prefer ONE comprehensive section over multiple sparse sections
- HARD LIMIT: Maximum 5 sections per chapter. Combine related subtopics into single sections to stay within this limit.
**CRITICAL**: The chapter's generationHint above describes what content this chapter should generate. If the generationHint references documents/images/data, then EACH section that generates content for this chapter MUST assign the relevant ContentParts from AVAILABLE CONTENT PARTS below.
NOTE: Chapter already has a heading section. Do NOT generate a heading for the chapter title.
## SECTION INDEPENDENCE
- Each section is independent and self-contained
- One section does NOT have information about another section
- Each section must provide its own context and be understandable alone
AVAILABLE CONTENT PARTS:
(No content parts specified for this chapter)
## CONTENT ASSIGNMENT RULE - CRITICAL
If AVAILABLE CONTENT PARTS are listed above, then EVERY section that generates content related to those ContentParts MUST assign them explicitly.
**Assignment logic:**
- If section generates text content ABOUT a ContentPart → assign "extracted" format ContentPart with appropriate instruction
- If section DISPLAYS a ContentPart → assign "object" format ContentPart
- If section's generationHint or purpose relates to a ContentPart listed above → it MUST have contentPartIds assigned
- If chapter's generationHint references documents/images/data AND section generates content for that chapter → section MUST assign relevant ContentParts
- Empty contentPartIds [] are only allowed if section generates content WITHOUT referencing any available ContentParts AND WITHOUT relating to chapter's generationHint
## ACCEPTED CONTENT TYPES FOR THIS FORMAT
The document output format (txt) accepts only the following content types:
table, bullet_list, heading, paragraph, code_block
**CRITICAL**: Only create sections with content types from this list. Other types will fail.
useAiCall RULE (simple):
- useAiCall: true → Content needs AI processing (extract, transform, generate, filter, summarize)
- useAiCall: false → Content can be inserted directly without changes (Format is "object" or "reference")
RETURN JSON:
{
"sections": [
{
"id": "section_1",
"content_type": "table",
"contentPartIds": ["extracted_part_id"],
"generationHint": "Description of what to extract or generate",
"useAiCall": true,
"elements": []
}
]
}
**MANDATORY CONTENT ASSIGNMENT CHECK:**
For each section, verify:
1. Are ContentParts listed in AVAILABLE CONTENT PARTS above?
2. Does this section's generationHint or purpose relate to those ContentParts?
3. If YES to both → section MUST have contentPartIds assigned (cannot be empty [])
4. Assign ContentPart IDs exactly as listed in AVAILABLE CONTENT PARTS above
IMAGE SECTIONS:
- For image sections, always provide a "caption" field with a descriptive caption for the image.
Return only valid JSON. Do not include any explanatory text outside the JSON.

View file

@ -1,73 +0,0 @@
TASK: Generate Chapter Sections Structure
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
CHAPTER: Hauptteil (Level 1, ID: chapter_2)
GENERATION HINT: Hauptinhalt der E-Mail, der die wesentlichen Informationen und Anliegen beschreibt.
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT sections: Focus on essential information only
- AVOID creating too many sections - combine related content where possible
- Each section should serve a clear purpose with meaningful data
- If no relevant data exists for a topic, do NOT create a section for it
- Prefer ONE comprehensive section over multiple sparse sections
- HARD LIMIT: Maximum 5 sections per chapter. Combine related subtopics into single sections to stay within this limit.
**CRITICAL**: The chapter's generationHint above describes what content this chapter should generate. If the generationHint references documents/images/data, then EACH section that generates content for this chapter MUST assign the relevant ContentParts from AVAILABLE CONTENT PARTS below.
NOTE: Chapter already has a heading section. Do NOT generate a heading for the chapter title.
## SECTION INDEPENDENCE
- Each section is independent and self-contained
- One section does NOT have information about another section
- Each section must provide its own context and be understandable alone
AVAILABLE CONTENT PARTS:
(No content parts specified for this chapter)
## CONTENT ASSIGNMENT RULE - CRITICAL
If AVAILABLE CONTENT PARTS are listed above, then EVERY section that generates content related to those ContentParts MUST assign them explicitly.
**Assignment logic:**
- If section generates text content ABOUT a ContentPart → assign "extracted" format ContentPart with appropriate instruction
- If section DISPLAYS a ContentPart → assign "object" format ContentPart
- If section's generationHint or purpose relates to a ContentPart listed above → it MUST have contentPartIds assigned
- If chapter's generationHint references documents/images/data AND section generates content for that chapter → section MUST assign relevant ContentParts
- Empty contentPartIds [] are only allowed if section generates content WITHOUT referencing any available ContentParts AND WITHOUT relating to chapter's generationHint
## ACCEPTED CONTENT TYPES FOR THIS FORMAT
The document output format (txt) accepts only the following content types:
table, bullet_list, heading, paragraph, code_block
**CRITICAL**: Only create sections with content types from this list. Other types will fail.
useAiCall RULE (simple):
- useAiCall: true → Content needs AI processing (extract, transform, generate, filter, summarize)
- useAiCall: false → Content can be inserted directly without changes (Format is "object" or "reference")
RETURN JSON:
{
"sections": [
{
"id": "section_1",
"content_type": "table",
"contentPartIds": ["extracted_part_id"],
"generationHint": "Description of what to extract or generate",
"useAiCall": true,
"elements": []
}
]
}
**MANDATORY CONTENT ASSIGNMENT CHECK:**
For each section, verify:
1. Are ContentParts listed in AVAILABLE CONTENT PARTS above?
2. Does this section's generationHint or purpose relate to those ContentParts?
3. If YES to both → section MUST have contentPartIds assigned (cannot be empty [])
4. Assign ContentPart IDs exactly as listed in AVAILABLE CONTENT PARTS above
IMAGE SECTIONS:
- For image sections, always provide a "caption" field with a descriptive caption for the image.
Return only valid JSON. Do not include any explanatory text outside the JSON.

View file

@ -1,73 +0,0 @@
TASK: Generate Chapter Sections Structure
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
CHAPTER: Schluss (Level 1, ID: chapter_3)
GENERATION HINT: Abschluss der E-Mail mit einer Zusammenfassung und einem freundlichen Gruß.
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT sections: Focus on essential information only
- AVOID creating too many sections - combine related content where possible
- Each section should serve a clear purpose with meaningful data
- If no relevant data exists for a topic, do NOT create a section for it
- Prefer ONE comprehensive section over multiple sparse sections
- HARD LIMIT: Maximum 5 sections per chapter. Combine related subtopics into single sections to stay within this limit.
**CRITICAL**: The chapter's generationHint above describes what content this chapter should generate. If the generationHint references documents/images/data, then EACH section that generates content for this chapter MUST assign the relevant ContentParts from AVAILABLE CONTENT PARTS below.
NOTE: Chapter already has a heading section. Do NOT generate a heading for the chapter title.
## SECTION INDEPENDENCE
- Each section is independent and self-contained
- One section does NOT have information about another section
- Each section must provide its own context and be understandable alone
AVAILABLE CONTENT PARTS:
(No content parts specified for this chapter)
## CONTENT ASSIGNMENT RULE - CRITICAL
If AVAILABLE CONTENT PARTS are listed above, then EVERY section that generates content related to those ContentParts MUST assign them explicitly.
**Assignment logic:**
- If section generates text content ABOUT a ContentPart → assign "extracted" format ContentPart with appropriate instruction
- If section DISPLAYS a ContentPart → assign "object" format ContentPart
- If section's generationHint or purpose relates to a ContentPart listed above → it MUST have contentPartIds assigned
- If chapter's generationHint references documents/images/data AND section generates content for that chapter → section MUST assign relevant ContentParts
- Empty contentPartIds [] are only allowed if section generates content WITHOUT referencing any available ContentParts AND WITHOUT relating to chapter's generationHint
## ACCEPTED CONTENT TYPES FOR THIS FORMAT
The document output format (txt) accepts only the following content types:
table, bullet_list, heading, paragraph, code_block
**CRITICAL**: Only create sections with content types from this list. Other types will fail.
useAiCall RULE (simple):
- useAiCall: true → Content needs AI processing (extract, transform, generate, filter, summarize)
- useAiCall: false → Content can be inserted directly without changes (Format is "object" or "reference")
RETURN JSON:
{
"sections": [
{
"id": "section_1",
"content_type": "table",
"contentPartIds": ["extracted_part_id"],
"generationHint": "Description of what to extract or generate",
"useAiCall": true,
"elements": []
}
]
}
**MANDATORY CONTENT ASSIGNMENT CHECK:**
For each section, verify:
1. Are ContentParts listed in AVAILABLE CONTENT PARTS above?
2. Does this section's generationHint or purpose relate to those ContentParts?
3. If YES to both → section MUST have contentPartIds assigned (cannot be empty [])
4. Assign ContentPart IDs exactly as listed in AVAILABLE CONTENT PARTS above
IMAGE SECTIONS:
- For image sections, always provide a "caption" field with a descriptive caption for the image.
Return only valid JSON. Do not include any explanatory text outside the JSON.

View file

@ -1,16 +0,0 @@
```json
{
"sections": [
{
"id": "section_1",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Einleitung mit Begrüßung und Vorstellung des Themas",
"useAiCall": true,
"elements": [
"Herzlich willkommen zu unserer Einführung in das Thema. In dieser Einleitung möchten wir Ihnen einen Überblick über die wichtigsten Aspekte und Ziele geben, die wir in den folgenden Kapiteln behandeln werden. Unser Ziel ist es, Ihnen ein umfassendes Verständnis für die Thematik zu vermitteln und Sie auf die kommenden Inhalte vorzubereiten."
]
}
]
}
```

View file

@ -1,26 +0,0 @@
```json
{
"sections": [
{
"id": "section_1",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Abschluss der E-Mail mit einer Zusammenfassung der wichtigsten Punkte.",
"useAiCall": true,
"elements": [
"Zum Abschluss möchten wir die wichtigsten Punkte unserer E-Mail zusammenfassen. Wir haben die aktuellen Entwicklungen und zukünftigen Schritte besprochen, um sicherzustellen, dass alle Beteiligten gut informiert sind. Sollten Sie weitere Fragen haben, zögern Sie bitte nicht, uns zu kontaktieren."
]
},
{
"id": "section_2",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Freundlicher Gruß zum Abschluss der E-Mail.",
"useAiCall": true,
"elements": [
"Wir danken Ihnen für Ihre Aufmerksamkeit und freuen uns auf eine erfolgreiche Zusammenarbeit. Mit freundlichen Grüßen, [Ihr Name]"
]
}
]
}
```

View file

@ -1,49 +0,0 @@
```json
{
"sections": [
{
"id": "section_1",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Einleitung und Kontext des Hauptinhalts der E-Mail",
"useAiCall": false,
"elements": [
"In diesem Abschnitt wird der Hauptinhalt der E-Mail vorgestellt, der die wesentlichen Informationen und Anliegen beschreibt. Ziel ist es, dem Empfänger einen klaren Überblick über das Thema und die wichtigsten Punkte zu geben."
]
},
{
"id": "section_2",
"content_type": "bullet_list",
"contentPartIds": [],
"generationHint": "Auflistung der Hauptpunkte oder Anliegen der E-Mail",
"useAiCall": false,
"elements": [
"Hauptanliegen des Absenders",
"Wichtige Informationen oder Daten, die mitgeteilt werden müssen",
"Fragen oder Anliegen, die geklärt werden sollen",
"Erwartungen oder gewünschte Handlungen des Empfängers"
]
},
{
"id": "section_3",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Detaillierte Beschreibung eines spezifischen Anliegens oder Themas",
"useAiCall": false,
"elements": [
"Ein spezifisches Anliegen oder Thema wird hier detailliert beschrieben, um dem Empfänger ein tieferes Verständnis zu ermöglichen. Dies kann beispielsweise eine Erklärung eines Problems oder die Darstellung einer Lösung sein."
]
},
{
"id": "section_4",
"content_type": "paragraph",
"contentPartIds": [],
"generationHint": "Zusammenfassung und Schlussfolgerung des Hauptinhalts",
"useAiCall": false,
"elements": [
"In der Zusammenfassung werden die wichtigsten Punkte des Hauptinhalts noch einmal hervorgehoben. Der Abschnitt endet mit einer Schlussfolgerung oder einem Aufruf zum Handeln, um den Empfänger zu einer Reaktion zu bewegen."
]
}
]
}
```

View file

@ -1,56 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_1
- Content Type: paragraph
- Generation Hint: Einleitung mit Begrüßung und Vorstellung des Themas
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (paragraph).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "paragraph",
"content": {{"text": "This is paragraph text."}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Following sections:
- section_1 (paragraph):
- section_2 (bullet_list):

View file

@ -1,56 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_1
- Content Type: paragraph
- Generation Hint: Einleitung und Kontext des Hauptinhalts der E-Mail
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (paragraph).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "paragraph",
"content": {{"text": "This is paragraph text."}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Following sections:
- section_1 (paragraph):
- section_2 (bullet_list):

View file

@ -1,59 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_2
- Content Type: bullet_list
- Generation Hint: Auflistung der Hauptpunkte oder Anliegen der E-Mail
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (bullet_list).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "bullet_list",
"content": {{"items": ["Item 1", "Item 2", "Item 3"]}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Previous sections:
- section_1 (paragraph):
Following sections:
- section_2 (bullet_list):
- section_3 (paragraph):

View file

@ -1,60 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_3
- Content Type: paragraph
- Generation Hint: Detaillierte Beschreibung eines spezifischen Anliegens oder Themas
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (paragraph).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "paragraph",
"content": {{"text": "This is paragraph text."}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Previous sections:
- section_1 (paragraph):
- section_1 (paragraph):
Following sections:
- section_3 (paragraph):
- section_4 (paragraph):

View file

@ -1,60 +0,0 @@
# TASK: Generate Section Content
LANGUAGE: Generate all content in DE language. All text, titles, headings, paragraphs, and content must be written in DE.
## SECTION METADATA
- Section ID: section_4
- Content Type: paragraph
- Generation Hint: Zusammenfassung und Schlussfolgerung des Hauptinhalts
- Target Output Format: TXT (accepted content types: table, bullet_list, heading, paragraph, code_block)
## CONTENT EFFICIENCY PRINCIPLES
- Generate COMPACT content: Focus on essential facts only
- AVOID verbose text, filler phrases, or redundant explanations
- Be CONCISE and direct - every word should add value
- NO introductory phrases like "This section describes..." or "Here we present..."
- Minimize output size for efficient processing
## INSTRUCTIONS
1. Generate content based on the Generation Hint above.
2. Create appropriate content that matches the content_type (paragraph).
3. The content should be relevant to the USER REQUEST and fit the context of surrounding sections.
4. Return only valid JSON with "elements" array.
5. No HTML/styling: Plain text only, no markup.
6. Keep content CONCISE - focus on substance, not length.
## OUTPUT FORMAT
Return a JSON object with this structure:
{
"elements": [
{
"type": "paragraph",
"content": {{"text": "This is paragraph text."}}
}
]
}
Output requirements:
- "content" must be an object (never a string)
- Return only valid JSON - no text before, no text after, no comments, no explanations, no markdown code fences
- Start with { and end with } - return ONLY the JSON object itself
- Generate meaningful content based on the Generation Hint
## USER REQUEST
```
entwirf eine email mit dem inhalt des files davor
```
## CONTEXT
## DOCUMENT CONTEXT
Previous sections:
- section_1 (paragraph):
- section_2 (bullet_list):
Following sections:
- section_4 (paragraph):
- section_1 (paragraph):

View file

@ -1,16 +0,0 @@
{
"elements": [
{
"type": "bullet_list",
"content": {
"items": [
"Anfrage zur Klärung eines offenen Punktes",
"Vorschlag für einen gemeinsamen Termin",
"Bitte um Rückmeldung bis Ende der Woche",
"Erinnerung an die bevorstehende Frist",
"Bereitstellung zusätzlicher Informationen auf Anfrage"
]
}
}
]
}

Some files were not shown because too many files have changed in this diff Show more