codexlens
by catlog22
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| CODEXLENS_TOP_K | No | Search result limit | 10 |
| CODEXLENS_DEVICE | No | Device (auto/cuda/directml/cpu) | auto |
| CODEXLENS_FUSION_K | No | RRF fusion k parameter | 60 |
| CODEXLENS_ANN_TOP_K | No | ANN fine search candidates | 50 |
| CODEXLENS_EMBED_DIM | No | Vector dimension (must match model) | 384 |
| CODEXLENS_FTS_TOP_K | No | FTS results per method | 50 |
| CODEXLENS_HF_MIRROR | No | HuggingFace mirror (e.g. https://hf-mirror.com) | |
| CODEXLENS_AUTO_WATCH | No | Auto-start file watcher after indexing | false |
| CODEXLENS_ANN_BACKEND | No | ANN backend (auto/usearch/faiss/hnswlib) | auto |
| CODEXLENS_EMBED_MODEL | No | Local fastembed model name | BAAI/bge-small-en-v1.5 |
| CODEXLENS_AST_CHUNKING | No | Enable AST chunking + symbol extraction | true |
| CODEXLENS_BINARY_TOP_K | No | Binary coarse search candidates | 200 |
| CODEXLENS_EMBED_API_KEY | No | API key for embedding API | |
| CODEXLENS_EMBED_API_URL | No | API base URL for embedding (overrides local model) | |
| CODEXLENS_INDEX_WORKERS | No | Parallel indexing workers | 2 |
| CODEXLENS_MAX_FILE_SIZE | No | Max file size in bytes | 1000000 |
| CODEXLENS_RERANKER_TOP_K | No | Results to rerank | 20 |
| CODEXLENS_EMBED_API_MODEL | No | Model name for embedding API | |
| CODEXLENS_MODEL_CACHE_DIR | No | Model download cache directory | |
| CODEXLENS_EMBED_BATCH_SIZE | No | Texts per API batch | 32 |
| CODEXLENS_EMBED_MAX_TOKENS | No | Max tokens per text (0=no limit) | 8192 |
| CODEXLENS_FIND_MAX_RESULTS | No | find_files result limit | 100 |
| CODEXLENS_LLM_EXPAND_MODEL | No | Model name for LLM query expansion | glm-5-turbo |
| CODEXLENS_RERANKER_API_KEY | No | API key for reranker | |
| CODEXLENS_RERANKER_API_URL | No | Reranker API base URL | |
| CODEXLENS_EXPANSION_ENABLED | No | Two-hop query expansion for NL queries | true |
| CODEXLENS_LLM_EXPAND_API_KEY | No | API key for LLM query expansion | |
| CODEXLENS_LLM_EXPAND_ENABLED | No | Enable LLM expansion globally for search_files | false |
| CODEXLENS_RERANKER_API_MODEL | No | Model name for reranker | |
| CODEXLENS_EMBED_API_ENDPOINTS | No | Multi-endpoint: url|key|model,... | |
| CODEXLENS_GITIGNORE_FILTERING | No | Recursive .gitignore filtering | true |
| CODEXLENS_LLM_EXPAND_API_BASE | No | API base URL for LLM query expansion | https://open.bigmodel.cn/api/paas/v4/ |
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 | |
Latest Blog Posts
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/catlog22/codexlens-search'
If you have feedback or need assistance with the MCP directory API, please join our Discord server