code-archaeology

Compound V is a multi-model orchestrator for Claude Code, layered on Superpowers: a deterministic router (Opus for risky work, Sonnet for the mechanical), cross-model review (Claude + Codex), 3 scouts (code · docs · domain). The pill that turns one freelancer into a squad.

Understand any GitHub repository before you clone it.

Codex skill for evidence-backed code archaeology from local git history

Seek traces of code left behind in Neovim's saved undo history. A read-only terminal explorer.

处女作 · 微信公众号爬虫（Playwright + Python）

Claude Code skill that investigates Chesterton's fence — point it at a weird line of code and it digs through blame, old commits, and PR archaeology to answer: why does this exist, and is it safe to delete? 🚧

AI-powered git history explainer - understand why code exists

Quill — ISA Spec Architect & Code Archaeologist | Bootable fleet agent twin with lighthouse keeper integration

Ask your Git history why code exists, with evidence: where each function was born, what broke it, who calls it. CLI, MCP server for AI agents, VS Code extension, GitHub Action. Local-first.

🔍 dig — generate a self-contained HTML code-archaeology report from any git repo. Single static Go binary per platform, stdlib only (no third-party deps); needs the git CLI on PATH. Beautiful output.

An AI-powered RAG tool that turns Git history into a searchable knowledge base to uncover the 'why' behind your code.

Read-only, structured access to a Git repository's history — as an MCP server and a CLI. Hardened against option injection and hostile repo config.

git blame for WHY, not who — a Claude Code skill that recovers the intent behind a line of code and refuses to invent one. The recover-side companion to git-why.

Legacy codebase excavation and architecture reverse-engineering engine detecting dead code and security smells in pure PHP

git blame that answers *why*, not just *who* — line-level git archaeology with no network and no configuration

Evidence-first code archaeology for Git history, pull requests, issues, reviews, and reverts.

Git archaeology for AI coding agents: distill a repo's full history into durable engineering memory (.repo-memory/).

git 이력을 추적해 '이 코드 지워도 되나'에 커밋 근거로 답하는 Claude Code 스킬. 포매터·스쿼시·코드 이동 노이즈를 넘어 진짜 도입 커밋을 찾습니다.

git blame tells you who. linelore tells you why: the PR, linked issues and review comments behind any line of code.

A coding-agent skill: run git blame, git log -L, and chase the linked issue/PR before explaining why code looks the way it does.