Work On Issue Loop

work-on-issue-loop

Drive an issue from "validated" to "PR reviewed to convergence" without stopping. Step 1 implements the issue and opens the PR; the review cycle after that is fix-pr-review-loop's convergence loop run against the new PR. fix-pr-review-loop owns the loop mechanics (bot selection, preflight, wait, stop conditions, cycle cap, resolution, terminal table). This skill adds the implementation step, the follow-on issue sweep, and a terminal state for a run that never produces a PR. With an existing PR and no issue to implement, use fix-pr-review-loop directly.

Input

• Nothing — the issue validated or planned this session; with none, stop and ask which issue. Never select an issue from a list.
• #<N> / <N> / URL / owner/repo#N.
• Optional targetBranch ({ issue, targetBranch } or a prose "target branch "): passed unchanged to work-on-issue. The review loop follows the PR's own base branch.

Steps

1. Implement, open the PR, and trigger the first review

Invoke the work-on-issue skill (Skill tool, skill: work-on-issue), passing the resolved issue number explicitly. It implements, verifies, commits, pushes, and opens the PR (Closes #<N>). It never requests review; that is this loop's job. Gate on its outcome:

• Stopped with no PR (closed issue, existing PR, wrong repo, or any other stop work-on-issue defines; it owns the list) → nothing to drive; stop and relay its report with its stop reason.
• PR opened → capture the PR number/URL and branch, then post the first trigger yourself as a separate one-line comment with no footer: gh pr comment <PR-number> --body "<band-derived trigger>". Do not wait on CI; the reviewer surfaces check failures itself. Match the repo's trigger phrase if recent PR comments show it differs.

Which trigger. validate-issue step 6 owns the first-review table with its Claude and Codex columns; this file states no boundary of its own. The score source is the issue just implemented: a stamped PR review: line in its Execution block overrides the band, else its [C<score>] title prefix. A missing score routes to the heaviest row.

Map a stamped model before posting it. claude.yml resolves only opus, sonnet and fable: a stamped sonnet or haiku posts @claude sonnet review, opus posts @claude opus review effort:high, fable posts @claude fable review effort:high, each keeping a stamped effort:<tier> in place of high. An unadmitted shorthand becomes the Action's route keyword and can select the write-capable fix-pr job, which pushes commits. On Codex, sonnet/haiku map to @codex luna review and opus/fable to the bare @codex review. fix-pr-review-loop step 1 owns both mappings.

Bot selection and preflight. Apply fix-pr-review-loop step 1's two rules: @claude by default, @codex only on explicit selection, never switch bots mid-cycle; before waiting, confirm a review workflow for the selected bot exists (with the Codex secret and variable checks when Codex is selected).

Record the trigger timestamp and set review_count = 1.

2. Run the convergence loop

Run fix-pr-review-loop steps 2 through 4 against the PR, carrying review_count and the trigger timestamp:

• fix-pr-review-loop step 2 — wait for the review to land (both jq gotchas, background until-loop, ~30-minute cap).
• fix-pr-review-loop step 3 — the merge-conflict check and the stop conditions: a bare LGTM with nothing left to fix stops the loop; when review_count > 5, stop at the first LGTM even with non-blocking findings; a Needs Updates verdict never stops the loop by cycle count alone. The one exception is the divergence brake: pr_cycle_count >= 4 and Needs Updates and every blocking finding sits in code an earlier cycle added. pr_cycle_count is fix-pr-review step 4's growth-check count read from the PR's trigger comments, never the in-memory review_count. A blocking finding that cannot be attributed to a cycle (no file:line, or a path the head no longer has) defeats the brake, and a line a base merge brought in is base-branch work. When the brake fires, stop and report per step 5's Diverging row.
• fix-pr-review-loop step 4 — resolve with fix-pr-review, increment review_count, loop to its step 2.

On a "Done" terminal state continue with step 3; on any other terminal state go to step 4.

3. File named follow-on issues (mandatory before reporting success)

On either "Done" state, sweep the PR body, commit messages, and any docs the diff changed for follow-on work the implementation named ("follow-on", "own issue", "future work", "next step", "not yet wired/deployed", "needs a follow-up"). This is separate from the review's Create Follow-up Issue section, which fix-pr-review handles.

• Search first (gh issue list --search "<keywords>" --state all); skip items that already have an issue.
• File each remaining item as a fully-specced issue per github-issue-format (complexity-prefixed title, complete body with a ## Plain simple English section under 55 words, attribution footer). Never file a stub; an item that cannot be specced yet is named in the step 4 report as deliberately unfiled.
• Include every filed issue URL and every deliberately unfiled item in the step 4 report.

4. Report

Report per fix-pr-review-loop step 5: same terminal-state table and same **Verification limitation:** handling (a Verification limitation line is not a finding; list each unverified source outside the word cap, omit when none). Three deltas:

• Replace its "PR was already merged/closed" row with: work-on-issue stopped with no PR → Nothing to drive. Relay its report with the stop reason it gave; zero review cycles ran.
• Add every follow-on issue filed in step 3 (URLs) and any item deliberately left unfiled.
• On the Diverging row, the scope yardstick is the acceptance criteria of the issue this loop implements; report which of them the PR already satisfies.

Red Flags — STOP

fix-pr-review-loop's Red Flags table applies while the loop runs. Additional rows:

Common Mistakes

fix-pr-review-loop's Common Mistakes list applies unchanged, including the cycle-cap rules its step 3 owns. One addition: skipping the step 3 follow-on sweep drops work named only in prose.