Codex Media MCP
by riqvip
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MEDIA_GID | No | GID for the container process. Default is 1000. | 1000 |
| MEDIA_UID | No | UID for the container process. Default is 1000. | 1000 |
| MEDIA_MCP_PORT | No | The host port for the MCP HTTP endpoint. Container always listens on 8765. Default is 8766. | 8766 |
| OPENROUTER_MODEL | No | The OpenRouter model to use. Default is google/gemini-3.5-flash-lite. The selected model/provider must support the media input modality. | google/gemini-3.5-flash-lite |
| MEDIA_HOST_FS_ROOT | No | Root filesystem path for media access. Defaults to / outside Docker. | / |
| OPENROUTER_API_KEY | Yes | Your OpenRouter API key. Required for media analysis. | |
| MEDIA_VIDEO_MAX_FPS | No | Maximum frame rate for temporary video uploads. Default is 2. | 2 |
| OPENROUTER_PROVIDER | No | Optionally pins an OpenRouter provider/endpoint slug and disables provider fallback. Leave empty to allow OpenRouter routing. | |
| MEDIA_REASONING_EFFORT | No | Controls thinking on compatible OpenRouter models. default leaves the provider policy unchanged; none disables thinking; low, minimal, medium, high, xhigh, and max request an effort level. Mutually exclusive with MEDIA_REASONING_MAX_TOKENS. | default |
| OPENROUTER_AUDIO_MODEL | No | Optional separate OpenRouter model for audio input. There is no automatic model fallback. | |
| MEDIA_REASONING_MAX_TOKENS | No | Set a positive token budget for thinking on compatible models. Mutually exclusive with MEDIA_REASONING_EFFORT; leave MEDIA_REASONING_EFFORT=default when using this. | |
| MEDIA_INSPECT_MAX_INTERVALS | No | Maximum number of intervals returned per check in inspect_media. Default is 100. | 100 |
| MEDIA_INSPECT_TIMEOUT_SECONDS | No | Timeout in seconds for local media inspection. Default is 900. | 900 |
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