ai-loops
llms.txt0 starsChanged 7 days agoMITView on GitHub
loops (github.com/arxdsilva/ai-loops) is an open-source multi-agent workflow plugin for Claude Code, GitHub Copilot, Codex, Cursor, Gemini CLI, OpenCode, Amp, Windsurf and Cline. It turns one ticket (GitHub, GitLab, Jira, Linear or plain text) into a pull request with proof: research, plan with a reviewer agent, parallel specialist engineers, an independent QA agent per engineer that re-tests every claim, a product reviewer that proves every acceptance criteria, and reflection. One master agent keeps the scope and asks the user only at…
In this repository35 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- skills/work/platforms/claude.mdCLAUDE.md
- clarifySkill
- consistencySkill
- debuggingSkill
- documentationSkill
- featureSkill
- hexagonalSkill
- initSkill
- jobs-and-state-machinesSkill
- lang-clojureSkill
- lang-cppSkill
- lang-csharpSkill
- lang-elixirSkill
- lang-erlangSkill
- lang-goSkill
- lang-javaSkill
- lang-kotlinSkill
- lang-phpSkill
- lang-pythonSkill
- lang-rubySkill
- lang-rustSkill
- lang-scalaSkill
- lang-swiftSkill
- lang-typescriptSkill
- memorySkill
- parallel-workSkill
- performanceSkill
- qualitySkill
- reportSkill
- review-integrationSkill
- statusSkill
- testingSkill
- workSkill
The file
Files
# loops > loops (github.com/arxdsilva/ai-loops) is an open-source multi-agent workflow plugin for Claude Code, GitHub Copilot, Codex, Cursor, Gemini CLI, OpenCode, Amp, Windsurf and Cline. It turns one ticket (GitHub, GitLab, Jira, Linear or plain text) into a pull request with proof: research, plan with a reviewer agent, parallel specialist engineers, an independent QA agent per engineer that re-tests every claim, a product reviewer that proves every acceptance criteria, and reflection. One master agent keeps the scope and asks the user only at opt-in checkpoints. A hard gate requires ≥70% coverage of new code and tests under the race detector. 15 backend languages. MIT license, one stdlib Go binary with zero dependencies. Install on Claude Code with `/plugin marketplace add arxdsilva/ai-loops` then `/plugin install loops@ai-loops` (or `npx ai-loops` from a terminal); run `/loops:init` and `/loops:work <ticket>` in a repository. Other platforms use `npx ai-loops --target <platform>`, the generated packages under `platforms/` or `install.sh --target <platform>`. Each rule is stated once in `specs/` and cited elsewhere. ## Docs - [README](https://raw.githubusercontent.com/arxdsilva/ai-loops/main/README.md): what loops is, quick start, comparison, FAQ - [Guide](https://raw.githubusercontent.com/arxdsilva/ai-loops/main/docs/GUIDE.md): install on every platform, updating, hooks, languages, parallelism, limits - [Workflow spec](https://raw.githubusercontent.com/arxdsilva/ai-loops/main/specs/WORKFLOW.md): phases, loops, checkpoints and decisions, `loops` CLI commands - [Roles](https://raw.githubusercontent.com/arxdsilva/ai-loops/main/ROLES.md): the 32 agent roles and who verifies whom - [Platforms](https://raw.githubusercontent.com/arxdsilva/ai-loops/main/specs/PLATFORMS.md): support tiers for each agent tool ## Rules - [Engineering discipline](https://raw.githubusercontent.com/arxdsilva/ai-loops/main/specs/ENGINEERING.md): make it work, then make it right (adapted from ub-mode) - [Testing](https://raw.githubusercontent.com/arxdsilva/ai-loops/main/specs/TESTING.md): new-code coverage and race policy gate - [Architecture](https://raw.githubusercontent.com/arxdsilva/ai-loops/main/specs/ARCHITECTURE.md): hexagonal dependency direction, checked by `loops arch` ## Optional - [Language packs](https://github.com/arxdsilva/ai-loops/tree/main/specs/languages): per-language gate checks, coverage formats and rule books - [Language research](https://raw.githubusercontent.com/arxdsilva/ai-loops/main/research/languages/README.md): the sourced research behind the packs - [AGENTS.md](https://raw.githubusercontent.com/arxdsilva/ai-loops/main/AGENTS.md): instructions for agents working on this repository
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

