aot-ship-loop
Ship a code change through the complete Art of Technology quality loop: committed feature branch, no-mistakes validation, lint and tests, pull request…
Skill by Art-of-Technology
Chalk Labs
Need help deploying aot-ship-loop?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/art-of-technology-aot-ship-loop
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: devops
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/aot-ship-loop
What's inside
- Skill name
aot-ship-loop- Description
- Ship a code change through the complete Art of Technology quality loop: committed feature branch, no-mistakes validation, lint and tests, pull request, Octopus Review to 4+/5, green CI, and optional post-merge container release verification. Use when the user asks to ship, gate, validate, run the full PR loop, follow no-mistakes plus Octopus, get CI passing, or prepare a change for safe deployment.
- License
- MIT
- Instructions
- 93 lines in SKILL.md · view on GitHub
Files (2)
- SKILL.md
- openai.yaml · agents/
SKILL.md (first lines)
# aot-ship-loop Move one scoped change from a feature branch to a verified release candidate. Compose the installed `no-mistakes` and `aot-pr-loop` skills when available; this skill controls their order and the hand-offs between them. ## Exit conditions Report success only when: - the intended change is committed on a feature branch; - no-mistakes completes its review, tests, docs, lint, push, PR, and CI gates; - the latest Octopus Review score is at least 4/5 with no unresolved critical finding; - every required PR check is green; and - if deployment was explicitly requested, the merged revision's artifacts are published and the deployed revision is verified. Stop at **ready to merge** unless the user explicitly authorized merging. Merge authority does not imply deployment authority. ## 1. Establish scope 1. Read repository instructions and inspect branch, status, diff, remotes, and recent commits. 2. Confirm the change is on a feature branch and contains only the requested scope. 3. Run the relevant targeted checks and commit the intended files. Never mix unrelated existing changes into the commit. 4. Write an intent that captures the user's outcome, constraints, compatibility promises, and accepted tradeoffs. Do not reduce it to a file list. Show the proposed commit or review-fix changes and obtain confirmation before committing when repository instructions require it. ## 2. Run no-mistakes Start the pipeline from the feature branch: ```bash no-mistakes axi run --intent "<complete intent>" ``` Add `--yes` only when the user has clearly granted unattended auto-fix consent. While a run is active: - let the pipeline own the worktree; - do not edit, commit, abort, restart, rebase, amend, or force-push; - inspect progress with `no-mistakes axi status`; - answer gates through the pipeline's printed `axi respond` command; and - accept pipeline-authored fix commits, then let it rerun the affected gates. Do not treat a local test pass as completion. Continue through review, tests, documentation, lint, push, PR creation, and CI. ## 3. Run the Octopus PR loop Once the PR exists, follow `aot-pr-loop` against that exact PR. 1. Fetch the PR, latest Octopus review, inline comments, and general comments. Prefer `gh-axi` over `gh` when available. 2. Parse the overall score, category scores, and every finding. 3. Read the referenced code and classify each finding as fix, false positive, or skip. 4. Before changing files, show the PR URL, scores, and proposed action for each finding; obtain user confirmation. 5. Apply the smallest valid fixes in new commits, push normally, react and reply to the relevant threads, then tag `@octopus` once for re-review. 6. Repeat until the latest score is at least 4/5. Never silently dismiss a critical finding. Inside this loop, never force-push, squash, rebase, or amend. If a pull conflicts, stop and surface the conflict. ## 4. Verify CI and hand off merge Wait for every required PR check. If a check fails, inspect its logs, make the smallest scoped fix with confirmation, rerun local validation, push a new commit, and return to the Octopus gate when the diff changed. When Octopus is 4+/5 and CI is green, report: - PR number, title, and URL; - no-mistakes outcome; - Octopus score and findings fixed or dismissed; - required CI checks; and - the exact next action: ready for user merge, or merge now if already authorized. ## 5. Optional merge and release verification Only after explicit authorization: 1. Merge using the repository's required strategy without rewriting history. 2. Wait for `main` CI to pass. 3. If containers are published, wait for **every** required container-image job. A green health endpoint alone does not prove the intended image is live. 4. Trigger a fresh deployment only when deployment was in scope. 5. Verify the deployed commit or image digest, plus one release-specific behavior. 6. Keep private services such as GPU, database, queue, and internal model endpoints private. If artifact publication or deployment verification fails, report the precise failed gate. Do not claim the release is live. ## Hard rules - No force-push, squash, rebase, or amend on a shared review branch. - No worktree edits while no-mistakes owns an active run. - No review-fix commit without the required confirmation. - No automatic merge or deployment without explicit authority. - No passing claim based only on health checks or stale CI. - No unrelated refactors or opportunistic cleanup.
From Art-of-Technology/skills, MIT.
More from Art-of-Technology/skills
- 0.529score
aot-nextjs-server-first
Art-of-Technology
Audit and refactor a Next.js app so data fetching happens on the server, backend credentials never reach the browser, and client-side fetch waterfalls stop…
SkillBFreshness: BClaude Code +9 - 0.529score
aot-ci-cd-optimization
Art-of-Technology
Measure and improve GitHub Actions CI/CD speed through independent parallel checks, cancellation of superseded runs, efficient secret scanning, caching and…
SkillBFreshness: BClaude Code +9 - 0.529score
aot-ci-workflows
Art-of-Technology
Cut wasted GitHub Actions runs in a repo and migrate it onto Art of Technology's shared reusable workflows (node-ci, dotnet-ci, docker-build, release), in two…
SkillBFreshness: BClaude Code +9 - 0.529score
aot-design-review
Art-of-Technology
Review a web UI against visual design, interaction, and accessibility principles, then produce a prioritized findings report with concrete code fixes.
SkillBFreshness: BClaude Code +9 - 0.529score
aot-release-lifecycle
Art-of-Technology
Enforce the trunk-based release lifecycle across environments.
SkillBFreshness: BClaude Code +9 - 0.529score
aot-security-audit
Art-of-Technology
Run a security audit on a TypeScript/Node.js or C#/ASP.NET Core codebase and produce a prioritized findings report with concrete fixes.
SkillBFreshness: BClaude Code +9 - 0.529score
aot-skill-audit
Art-of-Technology
Security audit of a third-party agent skill or plugin before anyone at Art of Technology installs it.
SkillBFreshness: BClaude Code +9 - 0.529score
blog-content-agent
Art-of-Technology
Research trends, analyze a product repository to learn what it does, propose blog topics, write a publication-ready post, optionally generate a banner image…
SkillBFreshness: BClaude Code +9 - 0.524score
actionable-output
Art-of-Technology
Make every response short, clear, and actionable.
SkillBFreshness: BClaude Code +9 - 0.524score
aot-pr-loop
Art-of-Technology
Address Octopus Review feedback on a GitHub pull request in a tight loop until the PR earns a 4+/5 quality score.
SkillBFreshness: BClaude Code +9 - 0.524score
aot-brief
Art-of-Technology
Shape agent replies for ADHD-friendly scanning and fewer output tokens.
SkillBFreshness: BClaude Code +9 - 0.516score
aot-rundown-current
Art-of-Technology
Give a concise recap of the current session and active task.
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Sep 1, 2026 (2 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/Art-of-Technology/skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from Art-of-Technology
- 0.529score
aot-nextjs-server-first
Art-of-Technology
Audit and refactor a Next.js app so data fetching happens on the server, backend credentials never reach the browser, and client-side fetch waterfalls stop…
SkillBFreshness: BClaude Code +9 - 0.529score
aot-release-lifecycle
Art-of-Technology
Enforce the trunk-based release lifecycle across environments.
SkillBFreshness: BClaude Code +9 - 0.529score
aot-ci-cd-optimization
Art-of-Technology
Measure and improve GitHub Actions CI/CD speed through independent parallel checks, cancellation of superseded runs, efficient secret scanning, caching and…
SkillBFreshness: BClaude Code +9 - 0.529score
aot-ci-workflows
Art-of-Technology
Cut wasted GitHub Actions runs in a repo and migrate it onto Art of Technology's shared reusable workflows (node-ci, dotnet-ci, docker-build, release), in two…
SkillBFreshness: BClaude Code +9 - 0.529score
aot-design-review
Art-of-Technology
Review a web UI against visual design, interaction, and accessibility principles, then produce a prioritized findings report with concrete code fixes.
SkillBFreshness: BClaude Code +9 - 0.529score
aot-security-audit
Art-of-Technology
Run a security audit on a TypeScript/Node.js or C#/ASP.NET Core codebase and produce a prioritized findings report with concrete fixes.
SkillBFreshness: BClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get aot-ship-loop running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
aot-ship-loop skill. Agent Mart. https://agentmart.dev/skills/art-of-technology-aot-ship-loop (last updated 2026-10-24)