# justrach/codedb

Zig code intelligence server and MCP toolset for AI agents. Fast tree, outline, symbol, search, read, edit, deps, snapshot, and remote GitHub repo queries.

Repository: https://github.com/justrach/codedb
Canonical: https://ross.abutalabs.com/products/codedb
Homepage: https://codegraff.com/blog/codedb-code-intelligence
Language: Zig
License: BSD-3-Clause
License Family: permissive
Topics: agentic-ai, agents, anthropic, cursor, gemini, mcp, openai, zig
Last push: 2026-08-26T17:15:22+00:00

## Health v2 (maintenance only)
Score: 78/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 87, longevity 13
- inputs: {"age_days": 183, "days_push": 7, "days_rel": 11, "gap_med": 0.0, "n_releases_24m": 65}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1367, forks 84 (observed 2026-08-28T04:04:31.297490+00:00)

## What it is
codedb is a Zig-based code intelligence server that indexes a codebase once and serves structural queries (symbols, outlines, trigram search, dependency graphs, snapshots) over MCP and HTTP. It acts as a fast context engine for AI coding agents like Claude Code, Cursor, and Codex, deliberately leaving editing to the agent's native tools.

## Use cases
- give my AI coding agent fast codebase search and symbol lookup
- index a large repo so agents stop shelling out to ripgrep
- query code structure like outlines, callers, and dependencies over MCP
- find symbols and definitions across a codebase instantly
- serve code context to Claude Code, Cursor, or Codex via MCP
- search a remote GitHub repository's code structure
- reduce token usage from unstructured grep output in agent workflows

## When to choose
- you run AI coding agents that repeatedly search large codebases and you want sub-millisecond indexed queries
- you need structural code intelligence (symbols, outlines, dependency graphs) exposed over MCP
- you want a zero-dependency single binary with file watching and incremental re-indexing

## When to avoid
- you need a full code editing or refactoring tool - codedb explicitly has no edit capability
- you need production-grade stability - it is alpha software with changing snapshot formats and parser gaps
- you need a network-accessible server - the HTTP server has no auth and binds to localhost only

## Facets
- artifact type: service
- maturity: experimental
- function: search-engine, parser, mcp, developer-tools
- domain: developer-tools, large-language-models, parsers
- platform: windows, cli, self-hosted
- tags: code-intelligence, mcp-server, code-search, symbol-index, trigram-search, dependency-graph, zig, context-engine, code-navigation, indexing, ai-agents, linux, macos

## Member repositories
- justrach/codedb (main) score 78

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:31.297490+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:41:07.571724+00:00, confidence not recorded.
  - readme: https://github.com/justrach/codedb (fetched 2026-08-28T04:04:31.297490+00:00, sha 7dfb2085542d)
  - homepage: https://codegraff.com/blog/codedb-code-intelligence (fetched 2026-08-29T11:58:12.648979+00:00, sha a2da64a55786)
  - site_page: https://codegraff.com/changelog (fetched 2026-08-29T11:58:12.658537+00:00, sha 74c7dc48bc45)
- Data as of 2026-08-30T08:39:29.467469+00:00.