CodeBase-MCP-Server CLAUDE.md

Project instructions for a code-search assistant that indexes a software repository and lets an AI assistant find relevant code through an MCP server, a standard connection for AI tools.

In plain words
What is it for?
It is for building and maintaining the repository index, code retrieval, MCP connection, and project conventions.
Why use it?
It gives the project a shared reference for its design, rules, and planned work, so changes stay consistent.

Instructions file

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add instructions/ayushshanu/codebase-mcp-server/claude-md
Clone the repo
git clone --depth 1 https://github.com/AyushShanu/CodeBase-MCP-Server
Per session 2,242 This file is loaded in full into every session.
When invoked 2,242 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.02242 $0.02242
Opus 5 $0.01121 $0.01121
Sonnet 5 $0.00448 $0.00448
Haiku 4.5 $0.00224 $0.00224

Measured yesterday against content hash 171e35a915b5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

CodeBase-MCP-Server CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 118 lines

How it starts

The opening of the file, as written. The whole thing — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Codebase RAG MCP Server — Project Memory

Overview

An AI developer assistant that indexes a real GitHub repository, understands code structure via Tree-sitter/AST-aware chunking, retrieves relevant code using hybrid BM25 + vector search, reranks the results, generates citation-backed answers, and exposes it all through an MCP server so Claude Code/Claude Desktop can use it. Ships open-source, MIT-licensed, clone-and-run — anyone can use it with only free API keys, no hosted backend required. This file is the source of truth for the roadmap, architecture, stack, and conventions — read it before starting any day's work. Full design rationale lives in Codebase_RAG_MCP_Final_Plan.md.

Constraints

  • Plan window: 20–31 August 2026 (12 days). Working V1 > polished V2 > extra features — cut scope, not the core retrieval + MCP demo, if behind.
  • Distribution: open-source, MIT license, bring-your-own-free-API-key. No hosted multi-user service, no remote-MCP-OAuth, no billing in this sprint.
  • Hardware: MacBook, Apple M5, 16GB unified memory, 512GB SSD. Local embedding/reranker/index footprint is well under 1GB — no concern. Optional local LLM mode: Qwen2.5-Coder-7B-Instruct at Q4_K_M (~4.5–5GB) fits comfortably. Never ship a large model crushed to Q2_XXS as the default — 2-bit quality loss undermines citation accuracy, which is the whole point of this tool.

Architecture

GitHub Repository
  → Repository Loader → File Filtering
  → Tree-sitter Parsing → AST-aware Code Chunks + file/line metadata
  → Embeddings (local)  ─┐
  → BM25 Index (local)  ─┤→ Hybrid Retrieval → Reranker (local)
                          ┘
  → LLM Answer Generation → File/Line Citations
  → MCP Server → Claude Code / Claude Desktop

Tech Stack — options are intentional, pick deliberately per spec

Layer Primary Alternative Alternative
LLM generation NVIDIA NIM Groq OpenRouter (+ optional Gemini, + optional local Qwen2.5-Coder-7B Q4_K_M for offline mode)
Embeddings (local only) all-MiniLM-L6-v2 (start here, Day 4) jina-embeddings-v2-base-code (swap before final) bge-small-en-v1.5
Reranker (local only) cross-encoder/ms-marco-MiniLM-L-6-v2 bge-reranker-base mxbai-rerank-base-v1
Vector store FAISS (faiss-cpu) Qdrant (local Docker or free cloud tier) Chroma
Keyword search rank-bm25
Parsing Tree-sitter (py-tree-sitter)
MCP Official MCP Python SDK

Read the full file on GitHub · 118 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 118 lines · 2,242 tokens per session scan A 171e35a915b5

Subscribe to this mod's changes

CodeBase-MCP-Server CLAUDE.md is an instructions file published in the GitHub repository AyushShanu/CodeBase-MCP-Server (1 stars, last pushed 3d ago), licensed MIT. It adds 2,242 tokens to every session, about $0.0112 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other instructions, from other repositories

spec-kit AGENTS.md

AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,104 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,182 tokens

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,345 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode · 5,001 tokens

next.js AGENTS.md

Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens