mcp-local-rag

Local-first RAG server for developers. Semantic + keyword search for code and technical docs. Works with MCP or CLI. Fully private, zero setup.

→ shinpr/mcp-local-rag · listed on github_topic

Is this your MCP?

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

github_topic

M8ven verifies MCPs across every public registry — install directly from whichever one you prefer.

// key findings

✅

No credential exfiltration, no sensitive file access, no obfuscation

Static analysis found nothing flowing your secrets to unexpected places.

✅

Open source with a license and README

Anyone can audit the code, the license is declared, and the publisher documents what it does.

// required environment variables

This server reads these from process.env. You'll be asked to provide them before it can run.

configBASE_DIR— Set to the folder you want to search (or BASE_DIRS for multiple roots — see [Configuration](#configuration)). Documents must live under one of the configured roots.

configBASE_DIRS— Set BASE_DIR to the folder you want to search (or for multiple roots — see [Configuration](#configuration)). Documents must live under one of the configured roots.

configCACHE_DIR— Profile model identifiers and quantization variants are fixed per release. Both profiles share the same (default: ./models/); the first run on each profile downloads its model.

configCHUNK_MIN_LENGTH— chunk-min-length 50 Minimum chunk length in characters (1–10000)

configCODEX_HOME

configDB_PATH— export =./my-db

configMAX_FILE_SIZE— max-file-size 104857600 (100MB) Maximum file size in bytes

configMODEL_NAME— ⚠️ CLI --model-name must match the MCP server's env var. Using a different embedding model against an existing database produces incompatible vectors, silently degrading search quality.

configRAG_DEVICE

configRAG_GROUPING— (not set) similar for top group only, related for top 2 groups.

configRAG_HYBRID_WEIGHT— 0.6 Keyword boost factor. 0 = semantic only, higher = stronger keyword boost.

configRAG_MAX_DISTANCE— (not set) Filter out low-relevance results (e.g., 0.5).

configRAG_MAX_FILES— (not set) Limit results to top N files (e.g., 1 for single best file).

configRUN_E2E

configRUN_E2E_KEEP_CACHE

// full audit trail

The full breakdown of what we checked, the deductions that landed, the network hosts, the dependency advisories, and concrete fix guidance is available to verified publishers.

// embed badge in your README

[![M8ven Score](https://m8ven.ai/badge/mcp/shinpr-mcp-local-rag-1y29hd)](https://m8ven.ai/mcp/shinpr-mcp-local-rag-1y29hd)

commit: 539d9df08b16a97be60013221aae890f25cb8c50

code hash: 216009af192e98726d42ab49e50e96a868b402e8dfb083bf976bb8e93694de16

verified: 6/4/2026, 11:31:24 AM

view raw JSON →