agy-delegate skill

Antigravity Delegate

You are the orchestrator. This skill lets you hand a bounded coding task to a separate implementer - the Google Antigravity CLI (agy) - then review what it produced and land it yourself. You write the brief and own the judgment; Antigravity does the typing in its own conversation; you verify and commit.

Nothing here is specific to one orchestrating agent. The loop needs only the ability to run a shell command and read a file, so any comparable agent can drive it. It is designed for and run on Claude Code; treat other orchestrators as designed-for, not yet proven.

When NOT to use this

the first-launch setup.

• The task is small enough to just do inline - delegation overhead is not worth it.
• The agy CLI is not installed or not authenticated. Install it from Antigravity's CLI docs and run

--read-only dispatch covers review without edits, but a plain review may not need delegation at all).

• You want to write the code yourself, or you only need Antigravity's opinion on code you wrote (a

Prerequisites (check once)

1. agy help succeeds. If not, install the Antigravity CLI and complete first-launch setup.
2. agy models succeeds. That proves the CLI can authenticate and list the available model labels.
3. You are in (or will point --cd at) the target git repository.

These checks do not prove that a headless write will be approved. In --print mode, Antigravity cannot prompt for a write permission and may auto-deny it. The relay detects that denial instead of reporting completion.

Choose the implementer model

agy has a configured default model, so --model is optional. Use it when the human has a preferred Antigravity model label for the task. Otherwise let Antigravity use its own current default rather than guessing.

The loop

Run these five steps per task. Steps 1, 4, and 5 are your judgment; 2 and 3 are mechanical.

1. Write the brief

Antigravity sees only the text you send plus what it can inspect in the workspace - no chat history, no shared context. Everything the task needs goes in the brief: the goal, the current state, what to change, what to leave untouched, the project's actual gate commands, and a report contract. Tell Antigravity it will not commit (you will). Keep one task per brief. Full guidance and a template: references/writing-the-brief.md.

2. Dispatch

Send the brief to Antigravity with the bundled helper. It wraps agy --print, captures the run, and writes a structured result.json - so your only job is "run a command, read a file." ( below is this skill's installed directory - the folder containing this SKILL.md.)

node "<skill-dir>/scripts/relay.mjs" --brief brief.txt --cd /path/to/repo
# choose a model label:                 add --model "<label from agy models>"
# reasoning effort (low, medium, high): add --effort high
# read-only (sandbox — no edits):       add --read-only
# enable Antigravity terminal sandbox:  add --sandbox
# resume the most recent conversation:  add --resume-last  (delta brief only)
# see all options:                      node .../relay.mjs --help

The helper starts a fresh Antigravity project by default and passes --add-dir (the --cd path, absolute) so agy has an explicit workspace. It does not pass --dangerously-skip-permissions by default. Mechanics, flags, and the result.json shape: references/dispatch-and-poll.md.

3. Wait for completion

The helper blocks until Antigravity finishes, so back it with whatever your orchestrator offers and resume when it returns:

the result file.

• Claude Code: run the Bash call with runinbackground: true; you are notified on completion.
• Plain shell / other agents: run it in the foreground for short tasks, or background it and poll

Do not trust progress trackers over reality: a run is finished when result.json is written and the process has exited. Read the working tree, not a status line. The implementer's full report is the finalMessage field in result.json (also printed in full on stdout between the report markers).

4. Review - do not trust the self-report

Antigravity's result.json includes its own final message and any gate claims. Re-verify, don't accept:

touchedFiles in the result is your starting point.

• Re-run the project's gates yourself (the test/lint/build commands from step 1).
• Read the diff against the brief: did Antigravity do what was asked, nothing more and nothing less?

• Run the relevant guard skills on the diff if you have them installed.
• For schema/migration changes, round-trip them; for removals, grep for dangling references.

Full checklist: references/review-and-land.md.

5. Land it

The implementer edits the working tree; the orchestrator commits. Only after the gates pass and the diff holds:

• Commit the verified work yourself, with a clear message.
• If it needs changes, send a delta brief with --resume-last and review again.

Permission model

Antigravity owns its own permission policy. The relay does not bypass it by default. Use --dangerously-skip-permissions only when the human explicitly accepts that Antigravity may auto-approve tool permission requests. --read-only composes --sandbox with --dangerously-skip-permissions: the sandbox is the enforcement — writes inside the workspace are overlaid and discarded, and paths outside it fail with EPERM — while the auto-approve only lets tools run inside it. As a user-facing flag it stays mutually exclusive with --dangerously-skip-permissions, which alone (without the sandbox) is full access. Use --sandbox on its own when you want the terminal sandbox enabled for a write run. If headless --print auto-denies a write, the relay reports status: "failed" and exits non-zero. The relay fingerprints the working tree before and after a --read-only run to report readOnlyViolation in result.json. Settings allow-rules do apply to headless --print runs, but their matching rules and config location vary by agy version and platform, so treat a denial as something to measure on your install rather than assume. Two traps have been measured. First, exact-match behavior: on Windows with agy 1.2.5, command() matched only a bare command with no arguments — command(git) never allowed a real git status, and command(regex:git .+) was required (issue #614 comment). On macOS with agy 1.2.0, that same bare command(git) rule did allow git status, so the exact-match trap is not a constant across versions. Second, on Windows, an open upstream defect in Antigravity's permission engine (issue #614) splits resolved paths on whitespace, so a binary under C:\Program Files\... — which is where git itself commonly lives, making this the single most common trigger for delegated coding tasks, not just node/npm — is matched as its first fragment and no command() rule can match it. On which file agy reads: verified live on Windows with agy 1.2.5, the effective rules were userSettings.globalPermissionGrants.allow in ~/.gemini/config/config.json, not the permissions.allow in ~/.gemini/antigravity-cli/settings.json that agy's own denial message points at; verified live on macOS with agy 1.2.0, rules in antigravity-cli/settings.json took effect. Recheck both on your version before relying on either file. See references/dispatch-and-poll.md for the full writeup. Do not add the bypass flag without explicit human approval.

Authorization model

Delegation is something the human opts into. Once they have ("run this queue", "proceed"), committing verified, gate-passing work is the agreed contract. Two limits on that mandate: surface, don't absorb (report Antigravity's design decisions, defensible-but-unasked turns, and non-blocking nitpicks rather than silently keeping them) and stop for scope changes (if correct completion needs going beyond the brief, ask - don't expand the mandate yourself). The full treatment is in references/review-and-land.md.

References

can execute blind: structure, XML blocks, the report contract, and real gate commands.

• references/writing-the-brief.md - how to write a brief Antigravity

result.json contract, backgrounding per orchestrator, and recovery when a run misbehaves.

• references/dispatch-and-poll.md - relay.mjs flags, the

boundary, and the rework cycle via --resume-last.

• references/review-and-land.md - the review checklist, the commit

carrying constraints forward, progress tracking, and the end-of-run coherence check.

• references/multi-task-queues.md - running a sequential queue: