sensegrep
by Stahldavid
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| AWS_REGION | No | AWS region used for Amazon Bedrock | |
| GEMINI_API_KEY | No | Gemini API key (AI Studio key) used for Gemini embeddings | |
| GOOGLE_API_KEY | No | Google API key alternative to GEMINI_API_KEY for Gemini embeddings | |
| OPENAI_API_KEY | No | OpenAI API key alternative for the OpenAI-compatible provider | |
| SENSEGREP_ROOT | No | MCP root directory (the project root the sensegrep MCP server operates on) | |
| SENSEGREP_WATCH | No | MCP watcher toggle | |
| FIREWORKS_API_KEY | No | Fireworks API key alternative for the OpenAI-compatible provider | |
| SENSEGREP_PROFILE | No | Named side-by-side index profile | |
| AWS_DEFAULT_REGION | No | AWS default region used for Amazon Bedrock | |
| SENSEGREP_PROVIDER | No | Embedding provider: one of `ollama`, `gemini`, `openai`, `bedrock`. Defaults to local Ollama when no API key or provider is configured. | |
| SENSEGREP_EMBED_DIM | No | Exact returned vector dimension for the embedding model, e.g. `1024` or `1536` | |
| SENSEGREP_EMBED_MODEL | No | Embedding model name, e.g. `qwen3-embedding:0.6b` (default), `gemini-embedding-001`, `qwen/qwen3-embedding-8b`, or `cohere.embed-v4:0` | |
| SENSEGREP_QUERY_CACHE | No | `true` by default; set `false` for controlled benchmarks or sensitive environments | |
| SENSEGREP_ANN_MIN_CHUNKS | No | Automatic ANN threshold; `0` disables | |
| SENSEGREP_BEDROCK_REGION | No | Amazon Bedrock region | |
| SENSEGREP_OPENAI_API_KEY | No | API key for an OpenAI-compatible embeddings provider | |
| SENSEGREP_BEDROCK_API_KEY | No | Amazon Bedrock bearer API key; omit when using the AWS SDK credential chain | |
| SENSEGREP_OLLAMA_BASE_URL | No | Ollama base URL (default `http://127.0.0.1:11434`) | |
| SENSEGREP_OPENAI_BASE_URL | No | OpenAI-compatible embeddings base URL (default `https://api.fireworks.ai/inference/v1`), e.g. `https://openrouter.ai/api/v1` or a local server's `/v1` base URL | |
| SENSEGREP_LANGUAGE_PLUGINS | No | Comma-separated list of project-relative ESM modules or package names providing language support plugins | |
| SENSEGREP_OPENROUTER_TITLE | No | Optional OpenRouter attribution header (title) | |
| SENSEGREP_EMBED_CONCURRENCY | No | Provider request concurrency for embeddings | |
| SENSEGREP_OPENAI_BATCH_SIZE | No | OpenAI-compatible request batch size | |
| SENSEGREP_OPENAI_CONCURRENCY | No | OpenAI-compatible provider request concurrency | |
| SENSEGREP_OPENROUTER_REFERER | No | Optional OpenRouter attribution header (referer) | |
| SENSEGREP_QUERY_CACHE_TTL_MS | No | Persistent query-vector cache TTL bound | |
| SENSEGREP_INDEX_EMBED_CONCURRENCY | No | Concurrent index embedding batches | |
| SENSEGREP_QUERY_CACHE_MAX_ENTRIES | No | Persistent query-vector cache maximum entries bound | |
| SENSEGREP_ADAPTIVE_HYBRID_DELAY_MS | No | Adaptive hybrid tuning delay | |
| SENSEGREP_ADAPTIVE_HYBRID_MIN_SCORE | No | Adaptive hybrid tuning minimum score |
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