forgeos
CLAUDE.md2 starsChanged 2 months agoMITView on GitHub
Entry point for Claude Code. It holds no rules of its own — it imports the operating contract so Claude Code and Codex obey the identical source of truth. @.ai/contract/core.md The import above is authoritative. Section 0 is the discovery gate and applies before any other section; section 2 says what else exists and which trigger loads it. Nothing on this page restates it. .claude/ is an adapter layer. It duplicates no rules — every file points into .ai/. .claude/README.md…
In this repository6 other files it gives its agents.
- AGENTS.mdAGENTS.md
- codebase-navigationSkill
- debuggingSkill
- documentation-updateSkill
- refactoringSkill
- test-generationSkill
The file
Files
CLAUDE.md
Entry point for Claude Code. It holds no rules of its own — it imports the operating contract so Claude Code and Codex obey the identical source of truth.
@.ai/contract/core.md
The import above is authoritative. Section 0 is the discovery gate and applies before any other section; section 2 says what else exists and which trigger loads it. Nothing on this page restates it.
Project Entry Points
- Project facts:
.ai/context/project.md - Hard constraints:
.ai/context/constraints.md - Active work:
.ai/tasks/active/,.ai/plans/active/
Claude Code Surfaces
.claude/ is an adapter layer. It duplicates no rules — every file points into .ai/.
| Surface | Location | Source |
| Subagents | .claude/agents/ | .ai/agents/ |
| Slash commands | .claude/commands/ | .ai/workflows/ |
| Skills | .claude/skills/ | .ai/skills/ |
| Permissions and hooks | .claude/settings.json | scripts/hooks/ |
.claude/README.md describes each surface and what it may contain. Permissions and hooks enforce a
subset of the contract mechanically — a safety net against accidents, not a substitute for the
contract and not a security boundary.
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.