imessage-mcp
by shadow173
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DEV | No | 1 for local development only: required variables may be omitted (a random password is printed at startup), the issuer may come from request headers, MODE defaults to mock, and the server listens on 127.0.0.1. Refused together with NODE_ENV=production. Never set it on a public host. | |
| HOST | No | Interface to listen on. | 0.0.0.0 |
| MODE | No | relay runs the tools on your Mac; mock serves seeded fake data. | |
| PORT | No | Port to listen on. | 3000 |
| TIERS | No | Only shapes the tool list served before the bridge first connects after a deploy. Match the bridge's TIERS; the gateway logs a warning when they differ. | read,send,org,pa,gpt |
| NODE_ENV | No | Optional. production makes the gateway refuse to start if DEV=1 is ever set by mistake; nothing else reads it. | |
| OWNER_TZ | No | IANA time zone, e.g. Europe/Berlin. Servers usually run in UTC, so set it to the same value as on the bridge. | the host's time zone |
| CIMD_HOSTS | No | Hosts allowed to serve OAuth client metadata documents. | chatgpt.com,claude.ai,claude.com |
| PUBLIC_URL | No | Public origin, with no trailing slash or /mcp. Must be https (plain http only for localhost). The OAuth issuer and resource come only from it. | |
| LINK_SECRET | No | At least 32 characters. Signs file and citation links. Same value on the bridge. Required in relay mode. | |
| REDACT_CODES | No | Hide verification codes. In mock mode, applies to the gateway. | true |
| SEND_ENABLED | No | Kill switch for everything recipients see. In mock mode, applies to the gateway. | true |
| BRIDGE_SECRET | No | At least 32 characters; the bridge presents it when it connects. Same value on the bridge. Required in relay mode. | |
| TOKEN_VERSION | No | Bump to sign out every client. | 1 |
| OWNER_PASSWORD | No | With DEV=1 only: a plaintext password instead of a hash. Ignored otherwise. | |
| SEND_ALLOWLIST | No | Allowlist for sends. In mock mode, applies to the gateway. | known |
| MOCK_AUTO_REPLY | No | Named contacts reply 15 to 30 seconds after you message them. In mock mode, applies to the gateway. | true |
| ACCESS_TOKEN_TTL_S | No | Access token lifetime, in seconds. | 3600 |
| ALLOW_INSECURE_CIMD | No | With DEV=1 only: allow non-HTTPS client metadata URLs. | false |
| AUTH_SIGNING_SECRET | No | At least 32 characters. Signs tokens and client ids. Rotating it signs everyone out and every connector must be added again. | |
| OWNER_PASSWORD_HASH | No | From npm run hash-password. | |
| REFRESH_TOKEN_TTL_S | No | Refresh tokens rotate on every use, so this is how long a client can sit unused before it must sign in again. | 2592000 |
| ALLOWED_REDIRECT_URIS | No | Extra exact OAuth redirect URIs. ChatGPT, Claude and loopback addresses are always allowed. | |
| SEND_RATE_LIMIT_PER_HOUR | No | Rate limit for sends. In mock mode, applies to the gateway. | 30 |
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 | |
This server cannot be deployed
Maintenance
ActivityMaintained
ResponsivenessNo issues