matrix-nio-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MCP_PORT | No | Port for the HTTP server | 8000 |
| QDRANT_HOST | No | Qdrant hostname (qdrant inside Docker Compose) | localhost |
| QDRANT_PORT | No | Qdrant port | 6333 |
| WEBHOOK_URL | No | OpenAI-compatible base URL for the LLM callback | |
| IGNORED_ROOMS | No | Comma-separated list of Matrix room IDs to exclude | |
| WEBHOOK_MODEL | No | Model name passed to the LLM | gpt-4o-mini |
| WEBHOOK_TOOLS | No | Optional JSON string of tools/parameters merged into the chat completions request body | |
| BACKFILL_LIMIT | No | Messages fetched per page per room during startup backfill | 100 |
| MATRIX_USER_ID | Yes | Full MXID, e.g. @bot:example.org | |
| OPENAI_API_KEY | Yes | OpenAI API key for embeddings | |
| EMBEDDING_MODEL | No | OpenAI embedding model | text-embedding-3-small |
| HTTP_AUTH_TOKEN | No | If set, requires Authorization: Bearer <token> for all endpoints | |
| MATRIX_DEVICE_ID | Yes | Device ID — must be stable across restarts for E2EE | |
| MATRIX_STORE_PATH | No | Path for the Olm E2EE crypto database (created if absent) | ~/.cache/nio-mcp/store |
| QDRANT_COLLECTION | No | Qdrant collection name | matrix_messages |
| SSE_QUEUE_MAXSIZE | No | Per-subscriber SSE event queue cap (oldest dropped when full) | 100 |
| ALLOW_SEND_MESSAGE | No | Set to true to enable the send_message tool | false |
| BACKFILL_PAGES_MAX | No | Maximum backfill pages per room; 0 = full history | 10 |
| MATRIX_ACCESS_TOKEN | Yes | Long-lived access token | |
| MESSAGE_BUFFER_SIZE | No | In-memory ring buffer size for get_recent_messages | 500 |
| WEBHOOK_BEARER_TOKEN | No | Bearer token sent in the Authorization header | |
| EMBEDDING_VECTOR_SIZE | No | Output dimension requested from the model and used for the Qdrant collection | 1536 |
| MATRIX_HOMESERVER_URL | Yes | Homeserver URL, e.g. https://matrix.example.org | |
| WEBHOOK_PROMPT_HEADER | No | Text prepended once before all per-message lines | New Matrix messages: |
| MATRIX_KEY_BACKUP_FILE | No | Path to an Element-exported E2EE key file; see Decrypting historical messages in encrypted rooms | |
| MATRIX_SYNC_TIMEOUT_MS | No | Matrix /sync long-poll timeout in milliseconds | 30000 |
| WEBHOOK_PROMPT_PER_MSG | No | Template rendered once per buffered message | {sender_name} ({sender}) in {room_name} ({room}): {message} |
| WEBHOOK_COOLDOWN_SECONDS | No | Seconds of silence before the LLM is called | 300 |
| MATRIX_KEY_BACKUP_PASSPHRASE | No | Passphrase chosen when exporting; required when MATRIX_KEY_BACKUP_FILE is set |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
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 | |
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/james-choncholas/matrix-nio-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server