DEVELOPER DOCUMENTATION

Build with multi-model intelligence.

Integrate the full 3-stage deliberation system into your applications via REST API or MCP server.

Authentication

All API requests require a Bearer token. Create API keys in your Settings page. Keys use the lc_ prefix.

Base URL: https://console.llm-counsel.com/api/v1/

REST API Endpoints

Query

POST /api/v1/query — Run a full 3-stage council deliberation. Pass a question (required) and optional council_id (UUID or template slug like "reasoning", "coders", "general").

Councils

• GET /api/v1/councils — List all councils (custom + templates)
• GET /api/v1/councils/{id} — Get a single council
• POST /api/v1/councils — Create a custom council (name, member_models, chairman_model required)
• PUT /api/v1/councils/{id} — Update a custom council
• DELETE /api/v1/councils/{id} — Delete a custom council

Models

GET /api/v1/models — List all available LLM models from the OpenRouter catalogue.

Conversations

• GET /api/v1/conversations — List all conversations
• GET /api/v1/conversations/{id} — Get full conversation with all messages and stages
• GET /api/v1/conversations/{id}/export — Download conversation as Markdown

Account

• GET /api/v1/me — Get your profile
• GET /api/v1/credits — Check your credit balance
• GET /api/v1/usage — Get recent queries and credit transactions

MCP Server

LLM Counsel is available as an MCP (Model Context Protocol) server for AI agents like Claude, Cursor, and other MCP-compatible clients.

SSE URL: https://console.llm-counsel.com/mcp/sse

Available tools: query_council, list_councils, create_council, update_council, delete_council, list_models, list_conversations, get_conversation, get_query_detail, check_credits, get_usage.

Error Codes

Status	Meaning
400	Bad Request — invalid input or validation errors
401	Unauthorized — missing or invalid API key
402	Payment Required — insufficient credits
403	Forbidden — accessing another user's resource
404	Not Found — council or conversation not found
422	Unprocessable Entity — empty question or invalid values
429	Too Many Requests — rate limit exceeded
500	Internal Server Error