An MCP server for efficient code indexing and symbol retrieval using tree-sitter AST parsing to fetch specific functions or classes without loading entire files. It significantly reduces AI token costs by providing O(1) byte-offset access to code components across multiple programming languages.
Claim it to get a verified publisher badge, a free copy of our full audit findings, and direct contact for any high-priority issues we find.
Install from
M8ven verifies MCPs across every public registry — install directly from whichever one you prefer.
Disclosed vulnerabilities in this server's declared npm dependencies (via OSV). Whether each is reachable depends on the installed versions.
bun vulnerable to OS Command Injection
Bun has an Application-level Prototype Pollution vulnerability in the runtime native API for Glo
process.env. You'll be asked to provide them before it can run.ANTHROPIC_API_KEY— export =sk-ant-...ASTLLM_LOG_FILE— Log to file instead of stderrASTLLM_LOG_LEVEL— warn Log level: debug, info, warn, errorASTLLM_MAX_FILE_SIZE_KB— 500 Max file size to index (KB)ASTLLM_MAX_INDEX_FILES— 500 Max files to index per repoASTLLM_PERSIST— 0 Persist the index to ~/.astllm/{path}.json after every index, and pre-load it on startup (1 or true to enable)ASTLLM_WATCH— 0 Watch working directory for source file changes and re-index automatically (1 or true to enable). Excluded dirs are never watched — see [Watch excludes](#watch-excludes).BINARY_PATHCODE_INDEX_PATH— ~/.code-index Index storage directoryGEMINI_API_KEYGITHUB_TOKEN— GitHub API token (higher rate limits, private repos)GOOGLE_API_KEY— export =...OPENAI_API_BASEOPENAI_API_KEYOPENAI_BASE_URL— export =http://localhost:11434/v1OPENAI_MODEL— export =llama3[](https://m8ven.ai/mcp/tluyben-astllm-mcp-a3yizy)