autopilot

autopilot is a command for coding agents from cirwel/unitares. It costs 12 tokens per session (436 once invoked), scanned A, original, Apache-2.0.

A command that runs predefined checks before handing work over or delivering it. It can inspect tests, unresolved Watcher results, branches, worktrees (separate working copies), and a proposed shipping step.

In plain words
What is it for?
Use it to run closeout diagnostics, choose when tests run, review branch and worktree hygiene, preview a shipping plan, or explicitly ship changes.
Why use it?
Final checks are easy to skip or reduce to automatic approval. This command reports whether work can proceed or still needs a person's attention while keeping delivery and destructive actions controlled.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/cirwel/unitares/autopilot
Clone the repo
git clone --depth 1 https://github.com/cirwel/unitares

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for autopilot

README.md
[![agentmods](https://agentmods.dev/badge/commands/cirwel/unitares/autopilot.svg)](https://agentmods.dev/commands/cirwel/unitares/autopilot)
Your own site
<a href="https://agentmods.dev/commands/cirwel/unitares/autopilot"><img src="https://agentmods.dev/badge/commands/cirwel/unitares/autopilot.svg" alt="Measured on agentmods" height="20"></a>
Per session 12 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 436 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00012 $0.00436
Opus 5 $0.00006 $0.00218
Sonnet 5 $0.00002 $0.00087
Haiku 4.5 $0.00001 $0.00044

Measured 4d ago against content hash f85935fd3a08, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

autopilot scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

commands/autopilot.md · 44 lines

What it actually says

Use this when the operator wants fewer rubber-stamp prompts and more explicit policy outcomes. The command bundles the boring loop while preserving the existing safety contract: diagnostics are automatic, delivery is explicit, and merge/destructive cleanup remains human-gated.

Default read-only run:

  • python3 scripts/dev/autopilot_closeout.py

Common options:

  • --test-policy auto|skip|always|staged|fresh controls test-cache.sh execution. auto skips docs-only changes and runs the cache for code-like changes.
  • --watcher-mode print|surface|skip prints unresolved Watcher output by default; surface mutates Watcher state from pending to surfaced.
  • --branch-hygiene includes the dry-run branch/worktree hygiene sweep.
  • --ship-plan "type(scope): message" previews the ship.sh route without mutating git state.
  • --ship "type(scope): message" explicitly runs ship.sh; add --stage-all only when the whole worktree belongs to the current task.
  • --json emits machine-readable output for wrappers.

Policy reading:

  • policy: proceed means diagnostics completed and no attention item remains.
  • policy: needs_human means automation did the bounded work, but judgment or explicit delivery remains. Typical causes are local-only changes, unresolved Watcher lines, repo-rooted processes, or branch hygiene holds.
  • policy: blocked means a command failed or closeout produced an error.

Rules:

  • Do not use --ship as a substitute for review. It creates the normal draft PR delivery artifact; merge/mark-ready remains deliberate.
  • Do not pass --stage-all if unrelated dirty files exist.
  • Do not stop processes, stash, boot out LaunchAgents, merge, force-push, or run destructive DB actions from this command.
  • Report the final delivery: line from the embedded workspace closeout.
Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 4d ago First seen · 44 lines · 12 tokens per session scan A f85935fd3a08

Subscribe to this mod's changes

autopilot is a command published in the GitHub repository cirwel/unitares (4 stars, last pushed today), licensed Apache-2.0. It adds 12 tokens to every session and 436 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.