Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add donnfelker/loop-skills/plugin install pr-autopilotWrote 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.
[](https://agentmods.dev/skills/donnfelker/loop-skills/pr-autopilot)<a href="https://agentmods.dev/skills/donnfelker/loop-skills/pr-autopilot"><img src="https://agentmods.dev/badge/skills/donnfelker/loop-skills/pr-autopilot/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/donnfelker/loop-skills/pr-autopilot"><img src="https://agentmods.dev/badge/skills/donnfelker/loop-skills/pr-autopilot.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00237 | $0.04642 |
| Opus 5 | $0.00118 | $0.02321 |
| Sonnet 5 | $0.00047 | $0.00928 |
| Haiku 4.5 | $0.00024 | $0.00464 |
Grade A, and why
pr-autopilot scanned grade A with 1 finding 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 9d 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.
Unrestricted tool accesslowExcessive agency
A wildcard tool grant or "run any command" leaves no least-privilege boundary at all.
targets, git hooks, or workflow files — which a malicious PR can rewrite to run arbitrary code in your Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
How it starts
The opening of the file, as written. The whole thing — 285 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PR Autopilot
Drive a single pull request's review feedback, its failing CI checks, and its merge conflicts to
completion without supervision. It runs round after round on its own — committing, pushing,
re-requesting review, and waiting for the next batch of feedback — until the reviews, the checks,
and the merge state all settle down. For a one-shot sweep instead of a loop, run it with
--single-pass: a single autonomous round, then stop.
Read ${CLAUDE_PLUGIN_ROOT}/references/pr-review-mechanics.md first (if that path doesn't
resolve, it's at the plugin root, two levels up from this skill:
../../references/pr-review-mechanics.md). That file is the shared "how to talk to the GitHub review
API correctly" layer — fetching all three comment sources, classifying, replying, resolving threads,
the per-bot re-review trigger table, (§7) inspecting CI check runs and reading their failure
logs, and (§8) checking mergeability and resolving merge conflicts. The steps below are this
skill's control flow: when to loop, when to commit, and when to stop.
Operating mode (read this first)
This skill runs fully autonomously — that's the whole point:
- No plan-approval gate. Still form a plan internally so your changes stay coherent, but don't stop for sign-off — the user has opted into unattended operation.
- You DO commit and push each round. Otherwise the loop can't make progress.
--single-passmode runs exactly one autonomous round (equivalent to--max-rounds 1) and then stops without waiting for the next batch — a one-shot "address everything outstanding right now" sweep, still with no approval gate. Everything else below is identical; the loop just doesn't re-enter.
Pushing commits and posting comments are outward actions, so keep them safe and faithful: only ever
push to the PR's own head branch, never force-push, and never post an "Addressed in <hash>" reply or
resolve a thread for a change you didn't actually make. The user can interrupt the loop at any time.
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.
- 9d ago First seen · 285 lines · 237 tokens per session scan A 3f1bd4118a75
pr-autopilot is a skill published in the GitHub repository donnfelker/loop-skills (18 stars, last pushed 1mo ago), licensed MIT. It adds 237 tokens to every session and 4,642 once invoked, about $0.0012 per session on Opus 5. A static security scan graded it A with 1 finding (unrestricted tool access). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
review-implement-phase
Implements triaged review actions, commits focused fixes, and posts Done plus resolves threads. Use when the user wants only the implementation phase of the review-framework workflow.
engram-branch-pr
PR creation workflow for Engram following the issue-first enforcement system. Trigger: When creating a pull request, opening a PR, or preparing changes for review.
verify-behavior
Verify or reproduce visible product behavior by driving the real UI with pi-computer-use's checked tools, requiring verified expect postconditions and durable state evidence for meaningful UI flows. Use when triage needs visual reproduction, implementation needs behavioral proof, review needs interactive confirmation…
github-contributor
End-to-end playbook for shipping high-quality pull requests to open-source projects you don't maintain — discovery, CONTRIBUTING compliance, PR-size check, minimal-diff implementation, PR description with AI-assisted disclosure, conflict resolution, and post-submission maintainer interaction. Use whenever creating…
revdiff
Review diffs, files, and documents with inline annotations in a TUI overlay, or answer questions about revdiff usage, configuration, themes, and keybindings. Opens revdiff in agterm/tmux/zellij/herdr/kitty/wezterm/cmux/ghostty/iterm2/emacs-vterm, captures annotations, and addresses them. Works in git, hg, and jj repos…
write-pr
Reference standards for writing pull request titles and descriptions in the tldraw repository, plus the pre-flight comment sweep over the diff. Use as supporting guidance when another skill or workflow needs PR content standards, not as the user-facing create/update PR workflow.