agentleFS
mia-cxmia-cx/skills/skills/autopilot/SKILL.md

autopilot

Skill4 starsChanged 28 days agoCC0-1.0View on GitHub

Use when the user says "autopilot": carry an issue or change unattended to a pull request that is ready to merge, or merged when they also say "ship it".

74 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use review-bugbot, rename-chat and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • 0mlkrizzz655335vSkill

    review-bugbot

    缺陷导向代码审查。用 Bugbot 子代理找 bug。

    Also in 3 projects that use this

  • 0mlkrizzz655335vSkill

    rename-chat

    重命名当前聊天。按聊天内容改标题。"disable-model-invocation: true

    Also in 3 projects that use this

  • Tsumugii24Skill

    split-to-prs

    Split current work into small reviewable PRs. Use when the user asks to split a chat, set of changes, branch, or PR.

    Also in 4 projects that use this

  • PandelisZSkill

    update-cli-config

    View and modify Cursor CLI configuration settings in ~/.cursor/cli-config.json. Use when the user wants to change CLI settings, configure permissions, switch approval mode, enable vim mode, toggle display options, configure sandbox, or manage any CLI preferences.

    Also in 3 projects that use this

  • 32103210Skill

    migrate-to-skills

    Convert 'Applied intelligently' Cursor rules (.cursor/rules/*.mdc) and slash commands (.cursor/commands/*.md) to Agent Skills format (.cursor/skills/). Use when you want to migrate rules or commands to skills, convert .mdc rules to SKILL.md format, or consolidate commands into the skills directory.

    Also in 3 projects that use this

  • 32103210Skill

    update-cursor-settings

    Modify Cursor/VSCode user settings in settings.json. Use when you want to change editor settings, preferences, configuration, themes, font size, tab size, format on save, auto save, keybindings, or any settings.json values.

    Also in 3 projects that use this

  • +1 more from 32103210
  • +1 more from 0mlkrizzz655335v

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 6 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 66 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Only an explicit "autopilot" grants this.

    line 15
  2. Distinctive:

    Prove: every change, per "Proof".

    line 31
  3. Distinctive:

    Gate: Check "The gate" against the live PR.

    line 34
  4. Distinctive:

    Run the checks the repo's CI runs.

    line 45
  5. Very specific:

    Add one focused test that fails without the fix, when the behavior has none and the testing skill calls for one.

    line 46
  6. Distinctive:

    All true on the latest pushed commit:

    line 53
  7. Distinctive:

    Ask only when you cannot continue alone:

    line 64

“Common” is measured: kept by at least 6 in 10 of the 66 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Autopilot

The user is not watching this thread. Your updates and final message go unread until long after you stop, so own the work from its current state to the finish without waiting to be prompted.

  • "autopilot" finishes at a PR that passes the gate, open and unmerged.
  • "autopilot, ship it", or an explicit merge request in the same ask, finishes merged.

Only an explicit "autopilot" grants this. Without it, each skill below keeps its own stopping point.

The grant

For this task only, the request authorizes:

  • Committing, pushing, rebasing, and --force-with-lease on the task's own branch.
  • Running the whole chain below without a go between steps. Where a skill would stop and report (address-issue at the filed PR, review-relay at its user gate, babysit at green), hand off to the next step instead.
  • Any test method Proof needs, including running the app, browser automation, and computer use, even where a project rule says to ask first. Rules about how to test still apply: the named command, fixture, browser, or check.
  • Merging through pr-merge, only with "ship it", and only below the high risk tier.

It grants nothing listed under "Asking the user", and it does not bypass a harness approval prompt.

Steps

  1. Build. An issue goes through address-issue, which files the PR. For any other change, branch off the default branch if you are on it, make the smallest change that completes the goal, and file it with pr-file. Fix the case that happened, not every case that could.
  2. Prove every change, per "Proof". Add one line to the PR description saying how.
  3. Review with review-relay, or review-council when the user asked for a council.
  4. Babysit with babysit until its Done holds. Prove each fix before pushing it.
  5. Gate. Check "The gate" against the live PR.
  6. Finish. Without "ship it", leave the PR open. With it, merge through pr-merge.
  7. Report: the PR link, final head and merge commit if any, your proof, each reviewer's verdict and score, every decision you made for the user, remaining risks, and whether the PR is open or merged.

Stay active from step 1 to step 6. Ending a turn while checks, bots, or delegates are pending stops the work, because nobody will prompt you again. Wait only on something that wakes you: a delegate's completion notification, babysit's --once watcher, or watch_pull_request in T3. Call unwatch_pull_request before you ask a question or stop.

Proof

Proof is evidence you observed yourself in this session: command output, a reproduction, a screenshot. Use the most direct proof the change allows:

  • Reproduce the original failure, apply the fix, and watch the same reproduction pass.
  • Run the checks the repo's CI runs.
  • Add one focused test that fails without the fix, when the behavior has none and the testing skill calls for one.
  • For UI and runtime behavior, run the app and look at it, because green mocked tests do not prove the experience works. UI PRs carry before-and-after captures, per address-issue.

A change you cannot prove is a change you are not confident in: ask.

The gate

All true on the latest pushed commit:

  • Babysit's Done holds.
  • Proof covers every change in the diff, review fixes included.
  • The review ended on a clean lap or round, not on budget, with no dispute left for the user.
  • Without "ship it": auto-merge is off and the PR is in no merge queue. Disable both before the first push, and report that you did.

Asking the user

A question in a normal message goes unseen, and the thread looks finished. Ask through the harness's question tool (AskUserQuestion in Claude Code, request_user_input in Codex) so the thread shows it is waiting. Put everything in the question: the PR link, what blocks you, and the options with your recommendation first.

Ask only when you cannot continue alone:

  • You cannot prove the change.
  • The gate cannot pass, and you cannot fix that yourself.
  • The next step is outside the grant: deleting data, touching production or live services, changing things outside the repo, closing a PR, or force-pushing another branch.
  • The goal is ambiguous in a way that changes the result.
  • Another PR makes this one obsolete.
  • A review dispute reached the user.
  • "Ship it" meets a PR whose Risk: line says high. Auth, billing, CI, release, and migration changes merge on the user's own approval.

Decide everything else with sensible defaults, and list those decisions in the report. If no question tool exists or it returns no answer, leave the PR open and unmerged, and start the final message with the question.

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.