multi-agent-docs
Skill2 starsChanged 16 days agoApache-2.0View on GitHub
Deterministically scaffolds and synchronizes CLAUDE.md, AGENTS.md, .claude/, and .agents/ using automated manifest inspection scripts followed by semantic LLM refinement. Use when configuring a repository for multi-agent workflows (Claude Code, Antigravity, Cursor, Codex) or when resolving instruction drift.
In this repository19 other files it gives its agents.
- .agents/AGENTS.mdAGENTS.md
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- cross-platform-dev-scriptsSkill
- lessons-learntSkill
- preflight-test-engineerSkill
- public-repo-release-reviewSkill
- readme-designerSkill
- saas-app-builderSkill
- third-party-skill-verifierSkill
- skill-creatorSkill
- cross-platform-dev-scriptsSkill
- lessons-learntSkill
- multi-agent-docsSkill
- preflight-test-engineerSkill
- public-repo-release-reviewSkill
- readme-designerSkill
- saas-app-builderSkill
- third-party-skill-verifierSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 3 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 36
Read the project source code and refine the generated CLAUDE.md and AGENTS.md files:
- Very specific: line 39
Read key modules, entry points, and tests (src/, lib/, tests/, etc.).
- Distinctive: line 44
Ensure CLAUDE.md and AGENTS.md have identical factual commands and architecture sections.
- Distinctive: line 52
Verify that both files are in parity and zero instruction drift occurred:
- Very specific: line 58
If validation passes ([PASS]), configuration is complete.
- Very specific: line 59
If warnings ([WARN]) or errors ([FAIL]) are returned for command discrepancies, resolve the drift in CLAUDE.md or AGENTS.md using …
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Multi-Agent Project Docs & Conventions
Purpose
Prevent instruction drift between tools by producing synchronized, high-accuracy CLAUDE.md and AGENTS.md configuration files alongside .claude/ and .agents/ project scaffolding through a deterministic code + semantic synthesis pipeline.
Procedural Workflow
Phase 1: Deterministic Scaffolding
Execute the built-in scaffolding script using run_command:
python <skill_path>/scripts/scaffold.py --target .The script automatically:
- Introspects repository manifests (
package.json,pyproject.toml,Cargo.toml,go.mod, etc.). - Extracts concrete build, test, lint, and run commands.
- Creates
.claude/,.agents/, and.agents/rules/directories. - Generates baseline synchronized
CLAUDE.md,AGENTS.md, and.agents/AGENTS.mdwith companion sync headers. - Emits a structured JSON summary of detected facts and remaining TODO items.
Phase 2: Semantic Content Synthesis & Refinement
Read the project source code and refine the generated CLAUDE.md and AGENTS.md files:
- Inspect Deep Project Context:
- Read key modules, entry points, and tests (
src/,lib/,tests/, etc.). - Identify domain conventions, error-handling patterns, and naming standards.
- Read key modules, entry points, and tests (
- Update Placeholders in Both Files:
- Replace all
[TODO: ...]placeholders with concrete architectural explanations. - Ensure
CLAUDE.mdandAGENTS.mdhave identical factual commands and architecture sections. - Preserve tool-specific sections:
CLAUDE.md: Claude Code subagent rules and instructions.AGENTS.md: Tool-agnostic rules for Antigravity, Cursor, and Codex.
- Replace all
Phase 3: Deterministic Sync Verification
Verify that both files are in parity and zero instruction drift occurred:
python <skill_path>/scripts/validate_sync.py --target .- If validation passes (
[PASS]), configuration is complete. - If warnings (
[WARN]) or errors ([FAIL]) are returned for command discrepancies, resolve the drift inCLAUDE.mdorAGENTS.mdusingreplace_file_content.
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.

