Geodata MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| SAM3_URL | No | URL of the separate SAM3 segmenter service. | |
| EMBED_DIM | No | Embedding dimension; defaults to 256. | 256 |
| EMBED_MODEL | No | Local embedding model name; choose a model with query/document prompts and output compatible with `EMBED_DIM`. | |
| VISION_MODEL | No | Vision model name for the OpenAI-compatible endpoint. | |
| VISION_API_KEY | No | Optional API key for the vision endpoint, if authentication is required. | |
| PUBLIC_BASE_URL | No | The viewer's reachable origin, so tool results contain usable map URLs. | |
| VISION_BASE_URL | No | OpenAI-compatible Chat Completions API base URL, without `/chat/completions`, for OCR, document questions and image comparison. | |
| GEODATA_API_KEYS | Yes | API keys for the MCP server; must contain at least one key (use two for the security tests). Generate keys with `openssl rand -hex 24`. | |
| VISION_EXTRA_BODY | No | Optional JSON object for endpoint-specific request fields. | |
| S3_PUBLIC_ENDPOINT | No | The reachable S3 endpoint, so tool results contain usable download URLs. | |
| VISION_CONCURRENCY | No | Number of concurrent vision requests; defaults to four. | 4 |
| VISION_OCR_EXTRA_BODY | No | Optional JSON object adding or replacing fields for transcription only. | |
| VISION_RESPONSE_FORMAT | No | Structured-output mode: `json_schema` (default), `json_object`, or `none` for servers without structured-output support. | json_schema |
| GEODATA_HTTP_CREDENTIALS | No | Per-host HTTP credentials for authenticated upstreams; credentials stay server-side. | |
| LANTMATERIET_CREDENTIALS | No | Credentials for authenticated Lantmäteriet upstreams; credentials stay server-side. | |
| VISION_MAX_OUTPUT_TOKENS | No | Maximum output tokens for the vision endpoint; can be raised to the model's limit. | 16384 |
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