kb-rag
by potanin
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| KB_RERANK | No | 0 skips the cross-encoder (saves the 1.1 GB download and some memory; a little less precise on vague questions). | 1 |
| KB_MCP_PORT | No | Change it in the service and in every client config together. | 8765 |
| KB_SKIP_DIRS | No | Directory names to skip inside repositories; replaces the default list. | .git,node_modules,build,... |
| KB_EMBED_MODEL | No | Any 768-dimensional fastembed model; for other sizes change EMBED_DIM in ingest.py and retrieve.py, then rebuild. | BAAI/bge-base-en-v1.5 |
| KB_CHUNK_TOKENS | No | Chunk size. Smaller chunks give sharper hits on code; larger ones keep more of an argument together in a paper. Change, then rebuild. | 450 |
| KB_CHUNK_OVERLAP | No | Chunk overlap. Smaller chunks give sharper hits on code; larger ones keep more of an argument together in a paper. Change, then rebuild. | 60 |
| KB_MCP_TRANSPORT | No | Transport for the MCP server. Set to sse and use /sse in the client URL to use the older SSE transport. | streamable-http |
| KB_MAX_FILE_BYTES | No | Files larger than this are skipped. | 2000000 |
| KB_MAX_PER_SOURCE | No | Chunks per file in one result set. Raise it when you ask about a single document. | 3 |
| TIKTOKEN_CACHE_DIR | No | Cache directory for tiktoken. Check it in ~/bin/kb and in the service if models download again. | |
| FASTEMBED_CACHE_PATH | No | Cache path for fastembed models. Check it in ~/bin/kb and in the service if models download again. |
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