pr-autoloop
Run PR maintenance to completion automatically after an agent creates a PR or a draft PR becomes ready for review: keep polling CI and bot review with gh, repair failures and reply in the repository's working language in the same worktree, escalate design loops, and squash merge when safe, without waiting for a human to restart each poll cycle. Use right after an agent-created PR exists or transitions draft-to-ready. Do NOT use for human-authored PRs, and do NOT redefine repair/merge/escalation policy here (that lives in pr-maintainer).
How do I install this agent skill?
npx skills add https://github.com/nextaltair/altairs-agent-dev-kit --skill pr-autoloopIs this agent skill safe to install?
- Gen Agent Trust Hubpass
This skill provides a standard mechanism for an agent to autonomously monitor GitHub pull requests, CI status, and bot reviews to complete maintenance tasks without human intervention between check cycles.
- Socketpass
No alerts
- Snykwarn
Risk: MEDIUM · 1 issue
What does this agent skill do?
PR Autoloop
Automatically drive the agent PR maintenance loop to a terminal state after a PR is created, without a human restarting each poll. This skill is independent of language, stack, and harness; only the non-blocking wait mechanism differs per agent (see "Per-Agent Wait Mechanism").
This is a bounded time-based loop with goal-based stop conditions. It polls an external system (GitHub PR state) on an interval, reacts to changed CI/review/mergeability state, and stops only on merge, escalation, or timeout. Do not use it as a generic autonomous coding loop; it exists for the well-defined PR maintenance workflow.
Relationship to pr-maintainer
This skill is the autonomous looping layer. It does not define policy.
- All policy - draft gates, review gate, repair rules, repair-loop limit (4), reply language, design escalation,
merge conditions, post-merge worktree cleanup - comes from the
pr-maintainerskill and the repository's PR maintenance ADR/policy, if one exists. - This skill only specifies the contract for how to keep the loop running by itself until a terminal state.
If anything here appears to conflict with pr-maintainer, pr-maintainer wins on policy and this
skill wins only on "do not stall waiting for a human between polls".
When to Use
Use immediately when any of these is true:
- This agent just created a PR (
gh pr create), including a draft created only because workflow says so. - A draft agent PR transitioned to ready for review (including a manual transition by the user).
- The user asks to continue/finish an agent-created PR through CI, review, repair, and merge.
First enforce the no-draft postcondition from pr-maintainer (ready the PR if it is reviewable), then
enter the loop. Do not stop at the PR URL when the PR is reviewable.
Autoloop Contract (agent-agnostic)
Treat maintenance as a self-paced poll loop. One loop cycle:
-
Gather state with
gh- all four checks are mandatory every cycle; never skip based on CI status:# (a) PR fields - isDraft, mergeStateStatus, reviewDecision, head SHA gh pr view "$PR" --repo "$REPO" \ --json number,isDraft,headRefOid,mergeStateStatus,reviewDecision,statusCheckRollup,labels # (b) Check run statuses gh pr checks "$PR" --repo "$REPO" \ --json name,state,bucket,link,startedAt,completedAt,workflow # (c) Bot reactions on the PR issue (Codex signals: +1=clean, eyes=in-progress) gh api --paginate "repos/$REPO/issues/$PR_NUM/reactions?per_page=100" # (d) Inline review comments - detect P1/P2 badge findings from Codex gh api --paginate "repos/$REPO/pulls/$PR_NUM/comments?per_page=100"$REPOisowner/repo(e.g.your-org/your-repo). On aghwithoutgh pr checks --json, read the checks fromstatusCheckRollupin (a).Steps (c) and (d) below are written for the Codex connector (
chatgpt-codex-connector[bot]). Which bot to expect, or that none is expected, comes from the "Review gate" inpr-maintainer: decide it before the first cycle. With another bot, substitute its login and its own "clean" signal. With no review bot, (c) and (d) still run every cycle to catch human comments, and the bot conditions in step 2 do not apply.mergeStateStatus=CLEANmeans GitHub merge preconditions pass; it does not mean bot review is complete. -
Classify the cycle into exactly one outcome:
-
continue - any of these: CI checks still running (step b has pending checks); no
chatgpt-codex-connector[bot]reaction yet on the PR issue (step c is empty or onlyeyes); CI is green but Codex review/reaction has not posted yet → schedule the next cycle. -
repair - any of these: CI checks failed (step b); Codex review state is
COMMENTEDand step d contains P1/P2 badge findings → fix in the PR worktree, validate, push, reply perpr-maintainer, then continue (count the repair against the limit). -
escalate - repair limit reached, recurring findings in the same boundary, or a forbidden change is required → open the design issue, comment on the PR, stop.
-
merge - all of the following are true (verified from step 1 data above):
- CI: all required checks pass (step b - no failed or pending).
- Draft:
isDraft == false(step a). - Codex reaction:
chatgpt-codex-connector[bot]has a+1reaction on the PR issue (step c).eyesalone → continue, not merge. - No blocking findings: step d has no unaddressed P1/P2 Codex comments.
⚠️
mergeStateStatus=CLEANwithreviewDecision=""alone does not satisfy this gate. In a repository with no review bot (per the Review gate), conditions 3 and 4 are replaced by: no unresolved change request or blocking comment from a human reviewer. → squash merge, clean up the worktree, stop. -
timeout - total elapsed reached the polling window with no terminal result → comment on the PR that CI/review did not complete in the window, stop.
-
-
If the outcome is continue, wait the poll interval and start the next cycle automatically.
The stop condition is not a prose judgment such as "the PR looks good". It is the conjunction of
observable predicates gathered through gh: CI is green, the PR is not draft, the expected bot review
artifact exists, no blocking findings remain, merge preconditions are clean, and the checked head SHA is
still current.
Loop parameters
- Poll interval: about 3 minutes between cycles.
- Polling window: at most 20 minutes total (about 6-7 cycles).
- Keep loop count and the last checked head SHA in session memory; track repair count with a todo.
- Never let the loop idle waiting for a human to manually re-trigger the next poll. The whole point of this skill is that "continue" reschedules itself.
Per-Agent Wait Mechanism
The only agent-specific part is how you wait the poll interval without blocking your runtime. Do not hardcode a single mechanism here; follow your agent guide:
- Claude Code → ScheduleWakeup self-pacing; fallback bounded
bash untilloop; neversleep && <next>. - Codex → inline session polling (no ScheduleWakeup). Keep the loop inside the active session and report the terminal state instead of relying on a platform routine to wake the agent later.
- Antigravity → the built-in
scheduletool (DurationSecondsfor the next cycle,MaxIterationsto bound the window). - OpenClaw → a one-shot automation addressed to the current session; do not wait with
sleep. - Any other harness → a scheduling primitive that wakes the same session if one exists; otherwise a bounded in-session poll. State which one was used.
導入先の CLAUDE.md / AGENTS.md に wait 機構のセクション (例: "pr-autoloop 実装") があれば
それを優先する。無ければ上のデフォルトに従い、どの機構を使ったか報告する。
Stop Conditions
Stop the loop (and report the final monitored state) on merge, escalate, or timeout. Do not merge
on continue, and do not treat an empty review/comment set as a clean review (it means "review not completed
yet" per pr-maintainer).
References
pr-maintainerskill - policy this skill executes.- The repository's PR maintenance ADR/policy doc, if one exists (for example under
docs/decisions/). - 導入先の
CLAUDE.md/AGENTS.md- per-agent wait mechanism のプロジェクト上書き (任意)。
How can the creator link this skill?
Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.
<a href="https://skillzs.dev/skills/nextaltair/altairs-agent-dev-kit/pr-autoloop">View pr-autoloop on skillZs</a>