Who Broke It — AI skill for Claude Code
Open-source coordination layer for AI-native engineering teams: one repo, many humans and coding agents.
How to install Who Broke It
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open shriyashish-mishra/who-broke-it and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Who Broke It does
Open-source coordination layer for AI-native engineering teams: one repo, many humans and coding agents. Task graph, intent and conflict detection, blast radius, verified handoffs, PR merge gate, git-native sync.
Alternatives in AI
- Paca — AI-native, free, open-source alternative to Jira, Trello, ClickUp & Monday 1.8k ★
- Real-World Pain Points — Project: https://github.com/Narwhal-Lab/MagicSkills For many teams building multi-agent systems, the first thi 277 ★
- Laravel AI Translator — High quality automatic translate your language files into many languages using AI like Claude, GPT and etc 250 ★
README
Who Broke It?
[](https://github.com/shriyashish-mishra/who-broke-it/actions/workflows/ci.yml) [](https://github.com/shriyashish-mishra/who-broke-it/releases/latest) [](LICENSE) 
One repo. Humans. AI agents. Zero excuses.
**The open-source coordination layer for AI-native engineering teams.** Your agents write the code. Who Broke It? (`wbi`) keeps the engineering from falling apart: a shared task graph, intent and conflict detection, blast radius, verified handoffs, pull requests built from them, and a merge gate. One static binary. No server. No account.
**[Website](https://shriyashish-mishra.github.io/who-broke-it/)** · **[Live dashboard](https://shriyashish-mishra.github.io/who-broke-it/demo-dashboard.html)** · **[Demo repo with real PRs](https://github.com/shriyashish-mishra/wbi-action-demo)** · [Docs](docs/) · [What is verified](docs/VERIFICATION.md)
$ wbi blame src/api/billing/
Last significant changes:
🤖 Claude TASK-007 1s ago feat(billing): status endpoint, add paused [wbi/TASK-007, unmerged]
🧑 Maya 3s ago initial commit
⚠ Contract BillingStatus changed by TASK-007
→ 3 downstream task(s) affected: TASK-008, TASK-009, TASK-016
Your team now has Claude Code, Codex, Gemini CLI, Cursor and a few humans all committing to one repo. The agents are good at writing code. Nobody is good at answering *who owns this, what depends on it, what did that agent just change that breaks mine, and is it actually done?* That is what this answers, from files that live in your repo.
**Status:** v0.2, pre-1.0. Proven end to end with
Related Skills
Living Codebase Cartographer
Offline codebase cartographer: language-agnostic repo maps, dependency/call graphs, blast-radius impact analys
Agents Kit
Conflict-free parallel development on one Git repo by multiple AI coding sessions (Claude Code / Codex). Git-o
Folder Lock
Many AI agents, one repo, no conflicting saves - a Claude Code skill: one lock per workfolder, handoffs instea
Project Desk
Shared workspace where Claude Code, Codex and humans coordinate on one codebase — an MCP server for task claim
Local Agent Workbench
A local-first macOS control plane for Claude Code, Codex and local AI with durable tasks, safe worktrees, huma
Codeblast
Know what breaks before you merge — mutation-tested code graph with architecture, change & impact maps. Eviden
Related Agents
Pan Previewer
Read-only foresight agent. Given a phase, set of phases, or milestone, produces a structured forecast (blast r
Workflow Knowledge Orchestrator
DM — Coordinate profiles, select workflows, decompose work, distribute context, track state, consolidate verif
Cloud Coordination Master
Use this agent when coordinating cloud and local operations, managing vault synchronization, handling conflict