開発ツール
probe
Semantic code search for AI — ripgrep speed with tree-sitter AST parsing.
★ 722 スターApache-2.0
ガイドを見る →cocoindex-codeは、cocoindex-ioが公開しているオープンソースの開発ツールプロジェクトです（GitHub: cocoindex-io/cocoindex-code）。GitHubスターは2,772、フォークは229で、Apache-2.0ライセンスの下で公開されています。GitHubでの最終更新は2026年10月6日です。
Lightweight AST-based semantic code search CLI (ccc) that makes coding agents faster and cheaper
Decide whether you want direct CLI search, or your coding agent to use semantic search automatically via the Skill/MCP integration.
Install cocoindex-code[full] with pipx or uv tool for local embeddings, or the slim build if you will use a cloud embedding API.
Optionally add the agent integration (npx skills add, or claude mcp add cocoindex-code -- ccc mcp), then let it initialize the project.
Run ccc index if needed, then ccc search "authentication logic" — or just ask your agent to find how something works.
Indexes code by syntax tree (tree-sitter), so search matches code structure and meaning, not just text.
The README emphasizes a 1-minute setup: install and go; ccc index auto-initializes new projects with defaults.
npx skills add cocoindex-io/cocoindex-code teaches Claude Code and other skill-compatible agents to initialize, index and search on their own.
ccc mcp runs a stdio MCP server exposing a search tool with refresh_index, language and path filters.
ccc grep finds code by structural example pattern against the syntax tree — no index or embeddings required.
init, index, search, grep, status, doctor, reset and a background daemon that keeps the index fresh.
Clone CocoIndex Code if you want the source; most users install the CLI in step 2.
git clone https://github.com/cocoindex-io/cocoindex-code.git
cd cocoindex-codeInstall the batteries-included build as the README recommends.
pipx install 'cocoindex-code[full]'
# or: uv tool install --upgrade 'cocoindex-code[full]'Initialize (optional), build the index, and run a semantic search.
ccc init
ccc index
ccc search "authentication logic"cocoindex-io maintains it, built on their CocoIndex engine.
Yes, Apache-2.0 licensed. The [full] install uses local embeddings with no API key; the slim install needs a cloud embedding provider key.
Not with the [full] install — local embeddings (default model Snowflake/snowflake-arctic-embed-xs) work out of the box. The slim install requires a cloud provider.
Text search finds exact strings. ccc adds AST-aware semantic retrieval with ranked chunks, file paths and line numbers, plus an MCP/Skill interface agents can call directly.
pipx install 'cocoindex-code[full]', run ccc index in your repo, then ccc search — or add the Skill so your agent does it for you.
Semantic code search for AI — ripgrep speed with tree-sitter AST parsing.
Fast, accurate semantic code search for agents — about 99% fewer tokens than grep plus read.
grepai: semantic code search and call graphs for AI agents, running 100% locally on your machine.
A codebase knowledge graph that makes Claude Code, Cursor, Codex and Gemini faster, cheaper and repo-aware.
Bridge local AI coding agents (Claude Code, Cursor, Gemini CLI, Codex) to messaging platforms (Feishu/Lark, DingTalk, Slack, Telegram, Discord, LINE, WeChat …
An MCP toolkit that gives coding agents IDE-grade semantic retrieval, editing and refactoring
LibreAIカタログに掲載された開発ツールプロジェクトの中で最もスターの多いもの — cocoindex-codeが合わなかった場合に検討したい、現実的な候補リストです。
The open source coding agent
The model-definition framework behind modern AI — thousands of pretrained transformers for text, vision, and audio.
Anthropic's agentic coding tool that lives in your terminal and works across your whole codebase
Cross-platform desktop all-in-one manager for Claude Code, Codex, OpenCode and other coding agents
OpenAI's lightweight coding agent that runs locally in your terminal
カタログの中でcocoindex-codeとスター数が近い開発ツールプロジェクト — 前後の棚もそのまま見て回れます。
A bridge between Streamable HTTP / SSE and stdio MCP transports — connect any client to any server
Symbol-level code retrieval MCP server (tree-sitter AST) built to slash agent token costs on exploration
Learn to code securely while having fun through our popular open source in-editor experience, designed for developers, students, and anyone curious about sec…
Claude Code skill: four-phase plan hardening (recon, interrogate, Codex adversarial review, cross-model build & inspection) — two AI models harden your plan …
One API for every agent harness — run Codex, Claude Code and more through a single self-hosted interface
A coding-agent harness optimized for smaller local LLMs — small models, serious coding work