agentleFS
Sign inStart free
DeL-TaiseiOzakiDeL-TaiseiOzaki/claude-code-orchestra/.agents/AGENTS.md

claude-code-orchestra

AGENTS.md199 starsChanged 2 months agoMITView on GitHub

This file is the Antigravity-specific adapter. The contract every CLI agent follows -- response structure, handoff rules, cross-CLI subagent invocation, and the completion-verification guardrails -- is the root AGENTS.md, which Antigravity loads automatically. Read that first. Antigravity operates as a Tier 1 (default) executor unless a task routes it elsewhere. See .claude/rules/tiers.md. agy -p is the non-interactive mode the shared wrapper drives: - Headless auto-approves every tool call. There is no caller-side flag that confines a run, which is why…

The file

Create a free account to mount
Files
AGENTS.md

AGENTS.md -- Antigravity Adapter

This file is the Antigravity-specific adapter. The contract every CLI agent follows -- response structure, handoff rules, cross-CLI subagent invocation, and the completion-verification guardrails -- is the root AGENTS.md, which Antigravity loads automatically. Read that first.

Antigravity operates as a Tier 1 (default) executor unless a task routes it elsewhere. See .claude/rules/tiers.md.

Headless Behaviour and Limits

agy -p is the non-interactive mode the shared wrapper drives:

python3 .claude/skills/_shared/cli_consult.py --cli antigravity --prompt-file <path>
  • Headless auto-approves every tool call. There is no caller-side flag that confines a run, which is why cli_consult.py --read-only is refused for this callee rather than accepted and silently ignored. A read-only Antigravity consultation is not available; use Codex with --sandbox read-only instead.
  • Text output is captured verbatim; the exit code is the only success signal (there is no JSON envelope as with claude -p).
  • --resume is not supported, so a multi-turn consultation must re-send its own context in the next prompt.
  • Because nothing confines the run, the wrapper's edits object and .claude/logs/cli-tools.jsonl are the record of what it changed. Pass --caller <your agent name> so a change traces back to the agent that asked for it.

Internal Context References

  • .claude/docs/DESIGN.md -- macro requirements and design decisions
  • .claude/rules/ -- coding standards, delegation, and tier rules
  • .claude/skills/ -- workflow skills and deterministic helpers
  • .claude/STATE.md -- active main agent and cross-session working state

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free