Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| EMBED_DIM | No | Embedding vector dimension | 1536 |
| LOG_LEVEL | No | DEBUG / INFO / WARNING | INFO |
| CHAT_MODEL | No | LLM for categorization + primer synthesis | gpt-4o-mini |
| GITHUB_PAT | No | GitHub Personal Access Token with repo scope | |
| DB_PASSWORD | No | PostgreSQL password (same as in DATABASE_URL) | |
| PG_POOL_MAX | No | asyncpg max pool connections | 10 |
| PG_POOL_MIN | No | asyncpg min pool connections | 1 |
| DATABASE_URL | Yes | PostgreSQL connection string | |
| MCP_TRANSPORT | No | FastMCP transport mode | streamable-http |
| OPENAI_API_KEY | Yes | OpenAI API key (for embeddings + LLM) | |
| EMBEDDING_MODEL | No | OpenAI embedding model | text-embedding-3-small |
| OPENAI_TIMEOUT_S | No | Per-request OpenAI timeout (seconds) | 60 |
| DEFAULT_LIST_LIMIT | No | Default result count for list_categories | 50 |
| GITHUB_BACKUP_REPO | No | Target repo in owner/repo format (e.g. isaacriehm/memory-backup) | |
| OPENAI_MAX_RETRIES | No | Exponential-backoff retry limit | 5 |
| DEFAULT_SEARCH_LIMIT | No | Default result count for search_memory | 10 |
| FASTMCP_JSON_RESPONSE | No | Set to 1 to force JSON responses | |
| BACKUP_INTERVAL_SECONDS | No | Seconds between backups (default: 21600 = 6 hours) | 21600 |
| PRIMER_UPDATE_MAX_AGE_S | No | Max seconds before auto primer regeneration | 3600 |
| MAX_CONCURRENT_API_CALLS | No | Semaphore for parallel OpenAI calls | 5 |
Capabilities
Server capabilities have not been inspected yet.
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
No tools | |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |