Spaces:
Sleeping
Sleeping
Multi-Agent Content Generation System - API Endpoints
Base URL: http://localhost:7860
Port: 7860 (all agents on single port)
Health & Status
Health Check
GET /health
Returns system health status and current run ID.
Response:
{
"status": "healthy",
"run_id": "uuid",
"port": 7860
}
Root Endpoint
GET /
Returns API information and available endpoints.
Pipeline Endpoints
Execute Full Pipeline
POST /api/pipeline/execute
Execute the complete 7-stage content generation pipeline.
Request Body:
{
"user_brief": "string",
"season_arc_document": "string",
"character_bible": "string",
"world_building_document": "string",
"character_voice_guide": "string",
"style_guide": "string",
"continuity_log": "string",
"hook_brief": "string (optional)"
}
Response:
{
"run_id": "uuid",
"status": "success",
"final_output": { ... },
"hf_output_url": "https://huggingface.co/...",
"hf_metadata_url": "https://huggingface.co/..."
}
Get Pipeline Status
GET /api/pipeline/status/{run_id}
Get the status and state of a specific pipeline run.
Response:
{
"run_id": "uuid",
"status": "completed|failed|running",
"pipeline_state": { ... }
}
Agent Endpoints
All agents run on port 7860 with dedicated endpoints.
1. Showrunner Agent
POST /api/agents/showrunner
Generate episode directive from user brief.
Input:
{
"user_brief": "string",
"season_arc_document": "string",
"character_bible": "string"
}
Output:
{
"episode_directive": "string",
"story_premise": "string",
"tone_brief": "string",
"character_focus_notes": "string"
}
2. Story Editor Agent
POST /api/agents/story-editor
Generate episode outline from directive.
Input:
{
"episode_directive": "string",
"series_continuity_log": "string",
"character_list": ["Alex", "Jordan", "Casey", "Morgan"]
}
Output:
{
"episode_outline": "string",
"act_structure": "string",
"story_notes_for_writers": "string"
}
3. Cultural Consultant Agent
POST /api/agents/cultural-consultant
Review outline for cultural accuracy.
Input:
{
"episode_outline": "string",
"world_building_document": "string",
"character_list": ["Alex", "Jordan", "Casey", "Morgan"]
}
Output:
{
"cultural_accuracy_notes": "string",
"flagged_inaccuracies": ["issue1", "issue2"],
"approved_touchpoints": ["touchpoint1"],
"reference_suggestions": ["suggestion1"]
}
4. Lead Writer Agent
POST /api/agents/lead-writer
Write episode script from outline and cultural notes.
Input:
{
"approved_outline": "string",
"cultural_consultant_notes": "string",
"character_voice_guide": "string",
"character_list": ["Alex", "Jordan", "Casey", "Morgan"],
"story_premise": "string"
}
Output:
{
"full_episode_first_draft": "string",
"scene_descriptions": "string",
"dialogue": "string",
"stage_directions": "string"
}
5. Dialogue Specialist Agent
POST /api/agents/dialogue-specialist
Polish dialogue in the script.
Input:
{
"first_draft_script": "string (must not be empty)",
"character_voice_guide": "string",
"character_list": ["Alex", "Jordan", "Casey", "Morgan"],
"dialect_slang_reference": "string"
}
Output:
{
"dialogue_polished_script": "string",
"voice_consistency_notes": "string"
}
6. Comedy Writer Agent
POST /api/agents/comedy-writer
Add humor and punch-ups to the script.
Input:
{
"dialogue_polished_script": "string (must not be empty)",
"hook_brief_from_showrunner": "string",
"character_list": ["Alex", "Jordan", "Casey", "Morgan"],
"tone_brief": "string"
}
Output:
{
"comedy_sharpened_script": "string",
"punch_up_notes": "string",
"hook_rewrite_for_opening": "string"
}
7. Proofreader Agent
POST /api/agents/proofreader
Final quality control on the script.
Input:
{
"comedy_sharpened_script": "string (must not be empty)",
"style_guide": "string",
"continuity_log": "string",
"character_list": ["Alex", "Jordan", "Casey", "Morgan"]
}
Output:
{
"final_locked_script": "string",
"qc_report": {
"issues": [],
"style_alignment": "string",
"character_consistency": "string",
"formatting_notes": "string",
"recommendations": []
},
"continuity_log_update": "string"
}
API Usage Examples
Example 1: Execute Full Pipeline
curl -X POST http://localhost:7860/api/pipeline/execute \
-H "Content-Type: application/json" \
-d '{
"user_brief": "A tech startup launches a new app",
"season_arc_document": "Season 1: The Rise",
"character_bible": "Alex (CEO), Jordan (CTO), Casey (Designer), Morgan (Marketer)",
"world_building_document": "Modern Silicon Valley",
"character_voice_guide": "Alex: sarcastic, Jordan: technical, Casey: perfectionist, Morgan: optimistic",
"style_guide": "Fast-paced, comedic, tech satire",
"continuity_log": "Episode 1: Team formation, Episode 2: First failure, Episode 3: Launch day"
}'
Example 2: Call Individual Agent
curl -X POST http://localhost:7860/api/agents/showrunner \
-H "Content-Type: application/json" \
-d '{
"user_brief": "A tech startup launches a new app",
"season_arc_document": "Season 1: The Rise",
"character_bible": "Alex (CEO), Jordan (CTO), Casey (Designer), Morgan (Marketer)"
}'
Example 3: Check Pipeline Status
curl -X GET http://localhost:7860/api/pipeline/status/0e35979c-8240-4451-a693-6bed1516aa4d
Error Handling
All endpoints return appropriate HTTP status codes:
- 200 OK - Successful request
- 400 Bad Request - Invalid input
- 404 Not Found - Resource not found
- 500 Internal Server Error - Server error
Error responses include a detail message:
{
"detail": "Error message explaining what went wrong"
}
Documentation
- Interactive API Docs:
http://localhost:7860/docs(Swagger UI) - Alternative Docs:
http://localhost:7860/redoc(ReDoc)
Architecture
Port 7860 (Single Port)
├── /health
├── /
├── /api/pipeline/
│ ├── execute
│ └── status/{run_id}
└── /api/agents/
├── showrunner
├── story-editor
├── cultural-consultant
├── lead-writer
├── dialogue-specialist
├── comedy-writer
└── proofreader
All agents communicate through the orchestrator and share state via the pipeline.