flow
CLAUDE.md1 starsChanged 5 months agoMITView on GitHub
Claude-Code-specific guidance. The shared, tool-agnostic project rules live in AGENTS.md, imported above and expanded into context automatically at session start. Each workflow stage in AGENTS.md is an explicit-invocation skill in .claude/skills/ (the Agent Skills open standard, so the same files are also picked up by Copilot CLI and other compatible tools). Invoking one creates the flow/ artifacts and, when given a tracker URL or work-item id, transitions the tracker state automatically (via .claude/scripts/tracker.sh; see docs/tracker-portability.md for the supported backends). Each…
Files
CLAUDE.md
@AGENTS.md
Claude-Code-specific guidance. The shared, tool-agnostic project rules live in AGENTS.md, imported above and expanded into context automatically at session start.
Workflow skills
Each workflow stage in AGENTS.md is an explicit-invocation skill in .claude/skills/ (the Agent Skills open standard, so the same files are also picked up by Copilot CLI and other compatible tools). Invoking one creates the flow/ artifacts and, when given a tracker URL or work-item id, transitions the tracker state automatically (via .claude/scripts/tracker.sh; see docs/tracker-portability.md for the supported backends). Each sets disable-model-invocation: true, so it runs only when you invoke it explicitly, never autonomously.
| Stage | Command | Artifact |
| Research (requirements) | /research-requirements | flow/research/ |
| Research (codebase) | /research-codebase | flow/research/ |
| Plan | /create-plan <issue URL> | flow/plans/ |
| Implement | /implement-plan <plan path> | code, tests |
| Validate | /validate-plan <plan path> | quality gate |
| Cross-vendor review (optional) | /cross-review [profile] | flow/reviews/ |
| PR description | /describe-pr | flow/prs/ |
| PR feedback | /handle-pr-feedback | code, plan updates |
See .claude/skills/ for the full set (including /commit, /iterate-plan, /create-handoff, and /resume-handoff). Run these on a feature branch and pause for manual verification between implementation phases.
Lessons learned
<!-- Add Claude-Code-specific lessons here as the project develops. -->In this repository15 other files it gives its agents.
- AGENTS.mdAGENTS.md
- commitSkill
- create-handoffSkill
- create-planSkill
- cross-reviewSkill
- describe-prSkill
- handle-pr-feedbackSkill
- implement-planSkill
- initSkill
- iterate-planSkill
- advanced-evaluationSkill
- research-codebaseSkill
- research-requirementsSkill
- resume-handoffSkill
- validate-planSkill
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.