[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M77MRW93XVM4JYJS0AEP45GK":3,"skill-same:01M77MRW93XVM4JYJS0AEP45GK":426},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":29,"commitSha":30,"skillDirectory":31,"forksCount":32,"repositoryHtmlUrl":33,"repositoryHomepageUrl":6,"ownerHtmlUrl":34,"ownerBlogUrl":35,"ownerDescription":36,"updatedAt":37,"install":38,"resources":50},"01M77MRW93XVM4JYJS0AEP45GK","pull-request",null,"When the bundled pr skill or built-in commit-push-pr command resolves in this session, prefer pr for a one-shot PR from committed work and commit-push-pr to commit, push and open one at once, only when no draft, body contract, or later ready, monitor, or merge step applies; this skill otherwise. Orchestrate the full PR lifecycle: prep (review + verify), create, ready (update branch, security review + verify, flip), monitor CI + review comments, merge, and fetch CI logs. Use when: 'create pr', 'ship it', 'pr prep', 'mark ready', 'ready for review', 'fix CI', 'address comments', 'monitor PR', 'merge this', 'check pr status', not for the all-PR babysit loop (use \u002Fbabysit-prs), branch\u002Fworktree lifecycle (use \u002Fworktree), or committing without a PR (use \u002Fcommit).","melodic-software\u002Fclaude-code-plugins","melodic-software","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F58273638?v=4","01M77JVQTXNGKCE4MDFY2BC936","Melodic Software","01M77JVQVN9QDWKAH6W03ANP76",30,"MIT","https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\ndescription: \"When the bundled pr skill or built-in commit-push-pr command resolves in this session, prefer pr for a one-shot PR from committed work and commit-push-pr to commit, push and open one at once, only when no draft, body contract, or later ready, monitor, or merge step applies; this skill otherwise. Orchestrate the full PR lifecycle: prep (review + verify), create, ready (update branch, security review + verify, flip), monitor CI + review comments, merge, and fetch CI logs. Use when: 'create pr', 'ship it', 'pr prep', 'mark ready', 'ready for review', 'fix CI', 'address comments', 'monitor PR', 'merge this', 'check pr status', not for the all-PR babysit loop (use \u002Fbabysit-prs), branch\u002Fworktree lifecycle (use \u002Fworktree), or committing without a PR (use \u002Fcommit).\"\nuser-invocable: true\ndisable-model-invocation: false\nargument-hint: \"[prep|create|ready|monitor|comments|ci|merge|reassess|evidence|status|full|…] [--unattended] [args]\"\nmetadata:\n  workflow-stage: pr\n  summary: Full PR lifecycle. Prep, create, monitor CI, address reviews, merge\n---\n\n**Arguments.** `[prep|create|ready|monitor|comments|ci|merge|reassess|evidence|status|full|…] [--unattended] [args]`. Full set: the Arguments table below. e.g., \u002Fpull-request prep, \u002Fpull-request create, \u002Fpull-request ready, \u002Fpull-request monitor, \u002Fpull-request comments, \u002Fpull-request merge, \u002Fpull-request evidence \u003Cpr> --evidence \u003Cfile>, \u002Fpull-request media \u003Cpr> --manifest \u003Cfile>, \u002Fpull-request status, \u002Fpull-request full, \u002Fpull-request fetch-logs \u003Cpr|run>, \u002Fpull-request view \u003Cpr> --diff, \u002Fpull-request list --head \u003Cbranch>, \u002Fpull-request review-venue\n\n## Repository context. Gather first\n\nThe read actions `view`, `list`, `review-comments` and `review-venue`, `comments --ci`, `conflicts --ci`, `dependency --ci` and `ci --ci` skip this section. For every other action, collect these with\n**individual** Bash calls, one command per call, never combined into a single invocation:\n\n- Current branch, `git branch --show-current`\n- Recent commits, `git log --oneline -5`\n- Working tree status, `git status --porcelain`\n- Changed files (staged+unstaged), `git diff --name-only HEAD`\n\nTreat a failure (not a repository, git unavailable) as an unknown value and carry on. Keep them\nseparate body calls, not pre-computed, so the skill stays invocable from inside a worktree: the composition record is the [gather-block convention](https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fdocs\u002Fconventions\u002Fgather-block\u002FREADME.md#the-pre-compute-block-runs-as-one-shell-invocation).\n**Pointer**: when an isolated session refuses a git command, fetch [How Claude Code enforces isolation](https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Fworktrees#how-claude-code-enforces-isolation) live.\n\n## Purpose\n\nRun the PR lifecycle from quality review through merge and the post-merge worktree transition, detecting state and resuming.\n\n**The two non-negotiable gates:**\n\n1. **Finding verification** (prep phase): agent review findings have a demonstrated error rate, so every finding is verified against current docs and actual code before it reaches the user.\n2. **Research-gated CI fixes** (monitor phase): no code fix without multi-source research consensus on the root cause. Unresearched \"obvious\" CI fixes are how wrong fixes ship.\n\n## Adapting to your environment (graceful degrade)\n\nThis skill is self-contained: it runs on `git`, the forge's CLI, `jq`, and its bundled scripts (skill-private ones under `${CLAUDE_PLUGIN_ROOT}\u002Fskills\u002Fpull-request\u002Fscripts\u002F`, plugin-shared ones under `${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002F`, which the `reference\u002F` files call `\u003Cscripts-dir>`: substitute the resolved path before a command from them reaches Bash). `jq` is required (the merge and readiness paths pipe forge API output through it): check `command -v jq` before the first phase that parses; if missing, stop with the install remediation (\u003Chttps:\u002F\u002Fjqlang.org\u002Fdownload\u002F>; a separate install under Git Bash on native Windows) instead of failing mid-phase. The readiness thread gate (Gate 7) also needs Python 3.11+; without it that gate holds as unproven and the rest of the skill runs. The `media` action and the stale reviewer-media check need Python 3.11+ too: without it `media` stops with the install remediation (\u003Chttps:\u002F\u002Fwww.python.org\u002Fdownloads\u002F>) and the stale check is skipped. An adjacent capability a phase names (a code-review skill or agents, a simplifier, a build\u002Ftest\u002Flint verifier, an external research skill, an exploration skill, a work-item tracker, a CI-log-audit agent, a forge-events push channel) is **optional**: invoke it when your environment provides it (a skill, plugin, agent, or MCP server), else follow the inline guidance, which stands on its own. Never block a phase because an adjacent tool is absent.\n\nConsumer conventions come from the consuming project's own `CLAUDE.md`, `AGENTS.md`, and rules. Notably: PR body template, branch naming, merge style (this skill defaults to squash), review-reply identity (some projects post bot-identity replies via a wrapper; default is the forge CLI under your own login), and any extra pre-PR gates. Read them before creating or merging.\n\n**Provider commands.** This body names forge operations (resolve the PR, read its checks, reply on a thread); their GitHub commands (the one forge supported today) are in [the GitHub provider reference](..\u002F..\u002Freference\u002Fproviders\u002Fgithub\u002FREADME.md). Read it before the first forge operation. The phase files under `reference\u002F` carry each phase's full GitHub procedure.\n\n**PR title format** resolves via the same ladder `\u002Fsource-control:commit` uses for the commit subject (see its SKILL.md), checked in order: the `pr_title_pattern` resolved across the layers of the source-control convention config (team value in `docs\u002Fconventions\u002Fsource-control.yaml`) per [..\u002F..\u002Freference\u002Fconfig-resolution.md](..\u002F..\u002Freference\u002Fconfig-resolution.md) → the consuming project's own `CLAUDE.md`\u002F`AGENTS.md`\u002Frules → Conventional Commits (11-type vocabulary. `build, chore, ci, docs, feat, fix, perf, refactor, revert, style, test`; `security` is not a member. **Pointer**: when the default set is questioned, fetch commitlint's [config-conventional `type-enum`](https:\u002F\u002Fgithub.com\u002Fconventional-changelog\u002Fcommitlint\u002Ftree\u002Fmaster\u002F%40commitlint\u002Fconfig-conventional#type-enum) live. **As of**: 2026-11-10. **Recheck trigger**: that rule's value changes) as the default. See [reference\u002Fcreate.md](reference\u002Fcreate.md) §2.4.1 for where the title is derived. When no layer resolves a `pr_title_pattern` and nothing is inferable, point the user at `\u002Fsource-control:setup`.\n\n**PR-body attribution** (the `🤖 Generated with [Claude Code]…` line) resolves from the `pr_body_attribution` key across the same layers (team YAML first, [..\u002F..\u002Freference\u002Fconfig-resolution.md](..\u002F..\u002Freference\u002Fconfig-resolution.md)). Absent → the default line, `none` → omitted, any other value → that line. It is the PR-body analogue of `\u002Fsource-control:commit`'s `trailer_policy` and gated separately; see [reference\u002Fcreate.md](reference\u002Fcreate.md) §2.4.1 for the assembly.\n\n**Branch-to-issue grammar** is a POSIX ERE whose LAST capture group is the work-item reference (an issue number or a tracker key), e.g. `^[^\u002F]+\u002F([0-9]+)-`. It resolves first from the `branch_issue_pattern` key across the layers (team value in the YAML; [..\u002F..\u002Freference\u002Fconfig-resolution.md](..\u002F..\u002Freference\u002Fconfig-resolution.md)), which `parse-branch-issue.sh` reads itself. The `branch_issue_pattern` userConfig is a deprecated fallback, read only when no layer sets the key: its value is `${user_config.branch_issue_pattern}` (this line is the substituted surface; a literal `${user_config.branch_issue_pattern}` token means unset), and [reference\u002Fcreate.md](reference\u002Fcreate.md) §2.4.0 fills its `\u003Cbranch-issue-pattern>` slot with a real ERE from it. When the script falls back to that value it prints a deprecation note on stderr; relay it to the user. Neither set: a bound work-item tracker's adapter supplies the grammar (§2.4.0), else the built-in `\u003Ctype>\u002F\u003CN>-\u003Cslug>` (and `routine-issue-\u003CN>`) convention.\n\n**PR open state** (`ready` or `draft`) decides whether `create` takes the new PR through `ready` or leaves it a draft. A draft means a person asked to see the work early or the PR is a multi-session container; ready means complete and through `ready`'s gates. The per-user value is `${user_config.pr_open_state}` (this line carries the substituted value; a literal `${user_config.pr_open_state}` token means unset). A `pr_open_state` key in the repository's `docs\u002Fconventions\u002Fsource-control.yaml`, read at the PR's base branch and never at its head, wins over it; a request for a draft wins over both; with none set the state is `ready`, and a consumer opts into `draft`. An unattended run follows the resolved value like an attended one, and `ready --unattended`'s holds still apply. [reference\u002Fcreate.md](reference\u002Fcreate.md) §2.4.3 holds the state meanings, resolves the state and reports the supplying level; the levels and the invalid-value rule are in [..\u002F..\u002Freference\u002Fconfig-resolution.md](..\u002F..\u002Freference\u002Fconfig-resolution.md).\n\n**Review venue** (`auto`, `local`, `ci` or `both`) decides which local review steps run before and at the ready flip. The per-user value is `${user_config.review_venue}` (this line carries the substituted value; a literal `${user_config.review_venue}` token means unset), passed as `--user` to `scripts\u002Fread-review-venue.sh`. A `review_venue` key in the repository's `docs\u002Fconventions\u002Fsource-control.yaml`, read at the PR's base branch, wins over it. [reference\u002Freview-venue.md](reference\u002Freview-venue.md) reads it and owns what each value does.\n\n**PR body preview** (`post` or `preview`) decides whether an attended run shows a body and waits for a yes before posting it. The per-user value is `${user_config.pr_body_preview}`; **PR body refresh** (`true` or `false`) decides whether a push to an open PR also rewrites the body sections it made stale, per-user value `${user_config.pr_body_refresh}`. A literal unexpanded token means unset. A key of the same name in the repository's `docs\u002Fconventions\u002Fsource-control.yaml` wins over each; the levels are in [..\u002F..\u002Freference\u002Fconfig-resolution.md](..\u002F..\u002Freference\u002Fconfig-resolution.md). [reference\u002Fcreate.md](reference\u002Fcreate.md) §2.4.3 applies the preview and [reference\u002Fmonitor.md](reference\u002Fmonitor.md) §3.3.2 the refresh.\n\n## Emit checklist\n\nFor runs spanning 3+ phases, copy `${CLAUDE_PLUGIN_ROOT}\u002Fskills\u002Fpull-request\u002Ftemplates\u002Fchecklist.md` into your project's working-notes location (or track it inline) and tick each `- [ ]` as its phase produces output. It survives `\u002Fclear`.\n\n## Arguments\n\n`$ARGUMENTS`. Action selector:\n\n| Action | Entry point | Use case |\n|--------|-------------|----------|\n| *(empty)* | Smart default | Detect current state, resume from right phase |\n| `prep` | Phase 1 | Review (skipped when the CI code-review lane covers it, [reference\u002Freview-venue.md](reference\u002Freview-venue.md)) + verify findings + simplify + comment pass + verify build |\n| `prep quick` | Phase 1 (fast) | Code errors only, skip simplify and the comment pass |\n| `prep review-only` | Phase 1 (partial) | Just review + verify findings |\n| `prep simplify-only` | Phase 1 (partial) | Just simplify + comment pass + re-verify |\n| `create [--draft '\u003Creason>']` | Phase 2 | Branch-name check + commit + changelog fragments where the repo releases from them ([reference\u002Fcreate.md](reference\u002Fcreate.md) §2.3.3) + push + open the PR as a draft, its body carrying the plan Brief's acceptance criteria when the branch implements a plan and `Closes` only for an issue whose done-when items the diff delivers, else `Refs:` with a `## Remaining` section ([reference\u002Fcreate.md](reference\u002Fcreate.md) §2.4.0, §2.4.1), then run `ready` on it when the resolved PR open state is `ready`; `--draft` requests a draft whose `Draft:` line names the reason. Reports the PR URL, its open state and the level that set it, then arms event delivery ([reference\u002Fcreate.md](reference\u002Fcreate.md) §2.6) when the session keeps acting on the PR, else stops |\n| `create --pushed --worktree \u003Cpath> [--draft '\u003Creason>'] [--hold '\u003Cline>']` | Phase 2 (PR-only) | **PR-only entry for an orchestrated flow**, the branch is already committed and pushed (by a dispatched worker), so this skips commit \u002F push \u002F branch update, re-resolves branch and diff from the given target worktree (not the session cwd), and runs body assembly + gates + opens the PR for `\u003Cbranch>`, then follows the resolved open state like an attended run (the repository level is read at the base branch, never from the worktree). `--draft` requests a draft whose `Draft:` line names the reason; `--hold` puts a `Hold:` line in the body the PR opens with. Used by `\u002Fwork-items:work`'s orchestrator after its pre-PR gate. See [reference\u002Fcreate.md](reference\u002Fcreate.md) §2.7 |\n| `ready` | Phase 2.5 | Update the branch from its base by the method the shared branch-update decision chooses (`ready --unattended` never rewrites a pushed head), run the code and security reviews over the PR diff in fresh-context subagents (each skipped when its CI review lane is present and healthy, [reference\u002Freview-venue.md](reference\u002Freview-venue.md)) and the verify gate on the updated head (its test suites skipped when CI's test lanes already passed on that head, reported as `skipped (CI covers it)`; `ready --verify-local` runs them anyway), flip the draft to ready (the base branch's `pre_ready` list can turn off any of its six configurable steps; each step's outcome is reported), post the `evidence` comment when an evidence file exists for the confirmed head (`--evidence \u003Cfile>` names one; attended runs only, and a refusal is reported, never undoes the flip), then hand the PR to `\u002Freview:explain-change`, whose `digest_policy` decides whether a digest is offered or built, then to `\u002Freview:reviewer-media`, whose policy decides whether a narrated walkthrough is built (neither gates anything). Never arms auto-merge: ready starts the AI reviews. Refuses on a branch with no PR, and leaves in draft a held or superseded draft, or one whose body closes an issue the diff does not finish. `ready --unattended` (every lane, loop or dispatched caller) also leaves a draft a person asked for (its `Draft:` body line) in draft. `ready [\u003Cpr>] --skip \u003Cstep>[,\u003Cstep>...] --reason \"\u003Ctext>\"` lets a person present skip configurable steps for this run only, reported and recorded in one PR comment, never in the body; refused unattended. See [reference\u002Fready-for-review.md](reference\u002Fready-for-review.md) |\n| `monitor` | Phase 3 | Watch CI, fix failures, evaluate comments. **Three-tier event delivery: (1) push channel** when your environment ships a forge-events channel (an MCP server delivering webhook events into the session), ~0 idle requests; **(2) Monitor tool** fallback (60s poll); **(3) polling** in cloud\u002Fheadless sessions, plus the PR-activity subscription where the session offers it. Check the push channel FIRST per [monitor.md](reference\u002Fmonitor.md) §3.0.05 before falling back |\n| `comments` | Phase 3.5 | Evaluate\u002Frespond to PR comments only |\n| `comments --ci` | Phase 3.5 (CI) | The pr-unblock lane's unattended mode: the runner's kept reviewer-bot threads, or an earlier pr-verify FAIL's notes, only, through its review-thread tool, running no head code. See [reference\u002Fcomments-ci.md](reference\u002Fcomments-ci.md) |\n| `conflicts --ci` | Phase 2.5.2 (CI) | Unattended conflict resolution for the pr-unblock lane: reads the runner's `conflicts.json` as data, resolves each conflicted path in place under the base-merge step's discipline, runs no command, never edits a path under `.github\u002Fworkflows\u002F`, and leaves a path whose two intents cannot both hold with its markers, which escalates it. See [reference\u002Fconflicts-ci.md](reference\u002Fconflicts-ci.md) |\n| `dependency --ci` | Phase 3.5 (CI) | The pr-unblock lane's unattended review of one risky Dependabot update: reads the context file, the checkout and the upstream release notes as data, leaves a scoped fix outside `.github\u002F` uncommitted for a later job to push, and returns a merge, fix, close or escalate verdict as structured output, running no head code. See [reference\u002Fdependency-ci.md](reference\u002Fdependency-ci.md) |\n| `ci --ci` | CI | The pr-unblock lane's unattended fix of a PR's failing checks: reads the runner's context file as data, reproduces each check with the repository's own command, fixes the cause, reruns it, and leaves the fix uncommitted for the runner's patch; never weakens a check, and escalates what it cannot fix with a named reason. See [reference\u002Fci-ci.md](reference\u002Fci-ci.md) |\n| `merge` | Phase 4 | Squash merge, or merge or enqueue without asking once the PR is `CLEAN` and the AI review checks have passed on the reviewed head, with the plain merge command (a refusal is reported, never handed to the user or retried in another shape), then confirm a merge-queue PR entered the queue (enqueue a `CLEAN` one left out, report the refusal text verbatim; §4.2.2) and wait for `MERGED` (no strategy flag behind a merge queue, `--auto` only pinned to the reviewed head; [reference\u002Fmerge.md](reference\u002Fmerge.md) §4.2.1) + worktree transition (reuse it, or release this session's lock) + verify. Never removes a worktree: that is `\u002Fsource-control:worktree cleanup` |\n| `evidence \u003Cpr> --evidence \u003Cfile> [--attach \u003Cpath>]... [--summary \u003Cfile>] [--artifact-url \u003Curl> [--retention-days \u003Cn>]]` | Evidence | After the verifier passed for the PR's head, post one evidence comment, or update it in place for a new head, with `gh pr comment --edit-last --create-if-none --attach`: the text summary, publishable captures only in collapsed sections laid out by `templates\u002Fevidence-comment.md`, and the CI capture artifact link with its expiry. A file outside the `publishable` list, or a leak-scan hit on the body, posts nothing. See [reference\u002Fevidence.md](reference\u002Fevidence.md) |\n| `media \u003Cpr> --manifest \u003Cfile> [--artifact-url \u003Curl> [--retention-days \u003Cn>]]` | Reviewer media | Place a produced walkthrough (video, contact sheet, chapter list, screenshots) in the marked `## Reviewer media` section near the top of the PR body, rewriting only the text between its two marker lines: claim the head first, attach only the files the producer's `manifest.json` lists with `gh pr edit --attach`, else link the Actions artifact with its expiry, and swap the request label for the posted one. A leak-scan hit, a screenshot with no frame-text sidecar, a claim another run holds, a body that changed under the edit, or no way to host the media posts nothing. `media stale \u003Cpr>` marks a section that shows an older commit; `media shows --facts \u003Cfile>` prints the commit it shows; `media fail \u003Cpr>` records a failed build once per commit. See [reference\u002Fmedia.md](reference\u002Fmedia.md) |\n| `reassess [\u003Cpr>] [--act] [--state-dir \u003Cdir>]` | Reassess | Judge whether one open PR should still land, even when green and mergeable, and return `continue`, `rework` or `close` with its evidence: report-only unless `--act` carries out the verdict. Runs on one PR and is no merge prerequisite. `--state-dir` defaults to `${CLAUDE_PLUGIN_DATA}\u002Fstate\u002Fbabysit-prs`. See [reference\u002Freassess.md](reference\u002Freassess.md) |\n| `status` | Report only | Unified status across all phases |\n| `full` | Phase 1-4 | Run prep → create → monitor → merge end-to-end |\n| `fetch-logs \u003Cpr\\|run> [--raw\\|--job \u003Cjob-id>]` | CI log retrieval | Pull failed-CI evidence: default = `::error`\u002F`::warning` annotations only (cheapest); `--raw` = full ZIP dump for archive review; `--job \u003Cid>` = per-job plain text |\n| `view [\u003Cpr>] [--repo \u003Cowner\u002Frepo>] [--diff] [--out \u003Cfile>]` | Read only | One PR's facts as JSON, or its diff with `--diff`; no number reads the current branch's PR. See [reference\u002Fread.md](reference\u002Fread.md) |\n| `list [--head \u003Cbranch>\\|--head-match \u003CERE>] [--search \u003Ctext>] [--state open\\|closed\\|merged\\|all] [--repo \u003Cowner\u002Frepo>]` | Read only | PRs as a JSON array, open by default |\n| `review-comments (--prs \u003Cn> \\| --pr \u003Cnumber>) [--repo \u003Cowner\u002Frepo>]` | Read only | The review comments of the last `\u003Cn>` merged PRs, or of one PR, as JSON lines. See [reference\u002Fread.md](reference\u002Fread.md) |\n| `review-venue [--repo \u003Cowner\u002Frepo>] [--max-age-days \u003Cn>] [--head \u003Csha>]` | Read only | Per AI review lane (code, security): `ci` when the repository's CI lane is present and healthy, so no local review runs for it, else `local`; with `--head`, a `tests` line: `ci` when CI's test lanes passed on that commit. See [reference\u002Freview-venue.md](reference\u002Freview-venue.md) |\n\nFor the all-PR continuous loop (discover every open PR, work each to readiness, self-pace), use\nthe sibling skill `\u002Fsource-control:babysit-prs`. It wraps this skill's per-PR review discipline\nin fleet orchestration and never merges.\n\n## Action defaults\n\n- **Merge mode:** `merge` squash-merges (one squashed commit per PR onto the default branch), see [reference\u002Fmerge.md](reference\u002Fmerge.md) §4.2. Follow the consuming project's convention when it differs. A stacked PR lands every layer below it: see [stacked-prs.md](..\u002F..\u002Freference\u002Fstacked-prs.md).\n- **Stack or separate:** before `create` opens a PR that builds on another open PR, decide stack versus sequential PRs off the trunk by \"When to stack\" in [stacked-prs.md](..\u002F..\u002Freference\u002Fstacked-prs.md).\n- **Monitor cadence:** `monitor` polls the PR's checks (the §3.0.1 REST read) and comment fetches every 60 seconds, see [reference\u002Fmonitor.md](reference\u002Fmonitor.md) §3.1; every loop follows the [API budget](..\u002F..\u002Freference\u002Fproviders\u002Fgithub\u002FREADME.md#api-budget). A wait is a REST poll that runs in the background (the §3.0.1 Monitor watch, or a `run_in_background` poll where Monitor is unavailable), never a foreground `sleep` or `until` loop and never the CLI's blocking checks watch, and it first reports each pending job as queued or running (\"Waiting on a pending check\"). A dispatched subagent does not wait: it stops once its PR is marked ready and reports the full head sha, and the dispatching session watches and merges ([reference\u002Fmonitor.md](reference\u002Fmonitor.md)).\n- **Required reviewers:** `create` requests no reviewers when it opens the PR. See [reference\u002Fcreate.md](reference\u002Fcreate.md) §2.4.3.\n\n## PR identity resolution\n\nPR identity (number, URL) resolves **live from the forge**, the authoritative source, at every phase that needs it. No caching, no state files.\n\n**Standard pattern** (every phase): resolve the current branch's PR number; when that is ambiguous (multiple open PRs, stale checkout, returning days later), name the branch explicitly. Both commands are in the provider reference's \"Resolve a pull request\".\n\nAfter resolving once at phase entry, **pass `\u003Cpr_number>` explicitly to all subsequent forge calls** in that phase (reading checks, merging, etc.). Never rely on bare-branch resolution mid-phase.\n\n**Why explicit PR numbers?** Resolution with no number goes by the HEAD branch, which fails when the worktree was cleaned up (branch context lost), multiple open PRs exist (wrong match), you return days later (stale checkout), or another session's PR merged first. Resolving once per phase bounds this risk to one point.\n\n---\n\n## Phase 0: Parse action and detect state\n\nParse `$ARGUMENTS` to extract the action (first token) and any sub-arguments, after taking out\n`--unattended` wherever it appears: it is the caller declaring that nobody is present. Pass it on\nto `ready` (which takes it as its own flag) and to no script; with nothing else left, the action\nis the smart default.\n\n**Smart default** (empty args): read live state from the forge to determine the right phase:\n\n```text\n1. Check git branch — on the default branch? → \"Create a worktree or branch first\"\n2. Resolve PR for current branch, reading its number, state and draft flag:\n   a. no PR found → no PR yet → START AT PHASE 1 (prep); skip steps 3-6\n      entirely (they all need a PR number that does not exist yet)\n   b. state = MERGED → skip to Phase 4.3 (worktree transition only: new branch from the default branch, delete the merged local branch)\n   c. state = CLOSED → report \"PR was closed without merging\" and stop\n   d. state = OPEN → capture pr_number, continue to step 3\n3. isDraft = true → START AT PHASE 2.5 (ready); skip steps 4-6. A draft owes its\n   review and its flip, not monitoring: no reviewer has been asked to read it yet\n4. Read the PR's checks — still running? → start at monitor\n5. Check for unaddressed comments → start at monitor (comments sub-phase)\n6. CI green + comments addressed → suggest merge\n```\n\nPresent the detected state and proceed to its phase, announcing it in interactive mode; in autonomous mode (`full`), proceed without pausing. Phase transitions are not decision points.\n\n**Status action**: read the PR's number, URL and state, then report current state across all phases.\n\n---\n\n## Phases\n\nExecute in order. Each phase is self-contained. Read the relevant file for detailed steps:\n\n| Phase | File | Entry actions |\n|-------|------|--------------|\n| 1. Prep | [reference\u002Fprep.md](reference\u002Fprep.md) | `prep`, `prep quick`, `prep review-only`, `prep simplify-only` |\n| 2. Create | [reference\u002Fcreate.md](reference\u002Fcreate.md) | `create` |\n| 2.5. Ready for review | [reference\u002Fready-for-review.md](reference\u002Fready-for-review.md) | `ready` |\n| 3. Monitor | [reference\u002Fmonitor.md](reference\u002Fmonitor.md) | `monitor`, `comments` |\n| 2.5.2 (CI) | [reference\u002Fconflicts-ci.md](reference\u002Fconflicts-ci.md) | `conflicts --ci` |\n| 3.5 (CI) | [reference\u002Fcomments-ci.md](reference\u002Fcomments-ci.md) | `comments --ci` |\n| 3.5 (CI) | [reference\u002Fdependency-ci.md](reference\u002Fdependency-ci.md) | `dependency --ci` |\n| CI | [reference\u002Fci-ci.md](reference\u002Fci-ci.md) | `ci --ci` |\n| 4. Merge | [reference\u002Fmerge.md](reference\u002Fmerge.md) | `merge` |\n| Evidence | [reference\u002Fevidence.md](reference\u002Fevidence.md) | `evidence` |\n| Reviewer media | [reference\u002Fmedia.md](reference\u002Fmedia.md) | `media` |\n| Reassess | [reference\u002Freassess.md](reference\u002Freassess.md) | `reassess` |\n\n---\n\n## Monitor entry\n\nOn `monitor`, `comments`, or `full` reaching monitor, read [reference\u002Fmonitor.md](reference\u002Fmonitor.md) before any CI polling or comment evaluation: its entry checklist runs first, and its per-iteration checklist governs every CI or comment event. Standalone `create` runs Steps 1-3 of that entry checklist when its session will keep acting on the PR it opened ([reference\u002Fcreate.md](reference\u002Fcreate.md) §2.6); `create --pushed` leaves them to its caller. `comments --ci` skips it, and `conflicts --ci`, `dependency --ci` and `ci --ci` do not enter Phase 3.\n\n---\n\n## Full lifecycle (`\u002Fsource-control:pull-request full`)\n\n`full` chains prep, create, monitor, and merge in one invocation. Read\n[reference\u002Ffull-lifecycle.md](reference\u002Ffull-lifecycle.md) when invoked with `full`, and only\nthen: it owns the phase-to-phase handoff state, the abort points, and what `full` does\ndifferently from running the four phases by hand. Any other action routes through the phase table\nabove instead.\n\n## Fetch CI logs (`\u002Fsource-control:pull-request fetch-logs \u003Cpr|run> [--raw|--job \u003Cjob-id>]`)\n\nPublic action for failed-CI evidence, a tiered fetch chain: cheapest signal first, escalating only when the lower tier is insufficient. Consumers say WHAT they want; the skill picks the internal helper from the flags.\n\n**Behaviors:**\n\n| Invocation | What it returns | When to use |\n|------------|-----------------|-------------|\n| `fetch-logs \u003Cpr-number>` (default) | `::error::` + `::warning::` annotations across all failed jobs | First-pass. Usually enough to identify the cause |\n| `fetch-logs \u003Cpr-number> --failed` | Annotations from failed jobs only | When the run has many jobs and noise is a concern |\n| `fetch-logs \u003Crun-id> --raw` | Full CI log archive, dumped in scope | When annotations are sparse \u002F missing. Debug-grade detail |\n| `fetch-logs \u003Crun-id> --job \u003Cjob-id>` | Plain-text log of one job | Targeted dive after seeing which job failed |\n\n`\u003Cpr-number>` and `\u003Crun-id>` are interchangeable; given a PR number, the skill resolves its latest run.\n\n**Composition with `monitor`:** `monitor` invokes this action on CI failure. Direct `fetch-logs` is for ad-hoc post-mortem (e.g., a closed PR's CI failure, auditing a green run for warnings).\n\n**Implementation note:** this action delegates to the bundled, private `fetch-annotations.sh` and `fetch-failed-logs.sh` scripts. Consumers MUST NOT cite script paths; use this action.\n\n## Read a pull request (`view`, `list`)\n\n`view` and `list` are the read-only actions other skills call instead of a forge CLI. Read\n[reference\u002Fread.md](reference\u002Fread.md) when invoked with either, and only then: it owns the\ncommand, the output fields, and the exit codes.\n\n## Read the review venue (`review-venue`)\n\n`review-venue` is the read action prep, ready and other skills call before a local AI review, so a\nreview runs in CI or locally, never in both. Read\n[reference\u002Freview-venue.md](reference\u002Freview-venue.md) when invoked with it, or when a phase\nreaches a review: it owns the command, what makes a lane healthy, and how each line is applied and\nreported.\n\n---\n\n## Important notes\n\n- **Side effects**: this skill commits, pushes, creates PRs, and merges. User approval gates at each dangerous step (commit message, CI fix, merge outside the [reference\u002Fmerge.md](reference\u002Fmerge.md) §4.2.1 merge after AI review) are the human checkpoints the skill enforces\n- **Finding verification is non-negotiable** (the first gate under Purpose)\n- **Research-driven fixes** are the point of the monitor phase: a short research burst costs little; an unresearched fix costs a lot\n- **Max 3 CI fix iterations**, to stop fix-push-fail loops\n- **Findings triage**:\n  - **Bot comments classified CORRECT** (Codex, claude-review, etc., after evidence-based verification against actual code): **auto-fix + test + react 👍 + D5 reply in the same turn, then push with the round's other fixes and post the D7 reply** ([monitor.md](reference\u002Fmonitor.md) §3.3.2). No user-approval pause: the CORRECT\u002FINCORRECT classification is the safety gate. Applies when the fix is small and scoped (\u003C~50 LOC, single concern); pause for cross-cutting refactors even when CORRECT\n  - **Bot comments classified INCORRECT**: autonomous 👎 reaction + reply with research-backed counter-evidence. Never silently ignore\n  - **Human reviewer comments**: always pause for user approval before reacting or fixing, regardless of classification\n- **Prep skips the review\u002Fsimplify work when no `code` class is detected**, and the verify gate reduces to lint. `ready` still reviews a diff that touches an instruction surface ([definition](reference\u002Fready-for-review.md#docs-only-and-instruction-surfaces)). Extra project-specific prep-evidence requirements come from the consuming project's own hooks\n\n---\n\n## Boundary, native Claude Code surfaces\n\nThree native surfaces cover parts of this lifecycle and get conflated with it when a PR is opened\nor watched.\n\n- **`pr` (bundled skill)**: creates one pull request generically; it gathers branch\n  context and applies Claude Code's own title, body, and attribution. The model and\n  the person can both invoke it.\n- **`\u002Fcommit-push-pr` (built-in command)**: commits, pushes, and opens a PR in one prompt-driven\n  step. The model and the person can both invoke it.\n- **`\u002Fautofix-pr` (built-in command)**: spawns a cloud session that watches the current branch's\n  PR and pushes fixes when CI fails or reviewers comment. Reserved for the\n  person to run; the model does not invoke it.\n- **This skill (marketplace plugin).** The whole lifecycle: prep with verified findings, a draft\n  under this repository's PR title and body contract, the ready flip after updating the branch from its base,\n  research-gated local monitoring with per-finding classification and replies, and merge.\n\n**Routing.** Where no draft discipline, body contract, or later lifecycle step applies: when the\nbundled `pr` skill resolves in this session, prefer it to open a PR from work already committed;\nwhen `\u002Fcommit-push-pr` resolves, prefer it only when the whole working tree belongs in the commit,\nsince it commits, pushes and opens the PR in one step. With unrelated uncommitted changes, `pr`\nis the native route, never `\u002Fcommit-push-pr`. Prefer\nthis skill whenever the repository declares a PR convention or the work continues into ready,\nmonitor, or merge.\n\n**Choose who watches the PR.** At the end of `ready`, and at `monitor` entry when no choice was\nmade, pick between this skill's monitor, a background agent, `\u002Fbackground`, `\u002Fautofix-pr` and the\nbabysit loop by whether the person is staying and whether the machine stays on, after reading the\nplan's rate-limit windows: at 95% or above, report the window and the options, and the person decides. For a route only\nthe person can start, print it ready to run. If \u002Fautofix-pr is available in your session (gate\nbasis: it registers gated and needs forge CLI and cloud-session access; records in\n[reference\u002Fnative-surfaces.md](reference\u002Fnative-surfaces.md)), the person can run it instead of or\nalongside this skill's monitor, with a filled-in prompt carrying this skill's review discipline. The\nmatrix and the prompt:\n[reference\u002Fwatch-handoff.md](reference\u002Fwatch-handoff.md). When the caller passes `--unattended`,\nwith any action, the run records that recommendation and the `\u002Fautofix-pr` prompt in its output\nand does not ask the person; without that flag it asks.\n\n**Mutation gate.** `pr` and `\u002Fcommit-push-pr` commit, push, and open a PR; `\u002Fautofix-pr` pushes\nfixes to the PR branch from a cloud session. This skill never chains into any of them on its own\nbehalf. When `\u002Fautofix-pr` runs alongside this loop, both push to one branch. Before each fix\ncommit, fetch the PR head and merge it into the checked-out branch (`git merge --ff-only` when there\nare no local commits, else `git merge`), then re-check the pending edit against the merged tree and\ndrop it when the cloud session already fixed that finding. A fetch alone leaves the local branch on\nthe old tip, and the push that follows is rejected as non-fast-forward.\n\n**Availability is never assumed.** `pr` and `\u002Fautofix-pr` register gated, and `\u002Fautofix-pr` also\nneeds forge CLI and cloud-session access; this section states what to do when a surface resolves, never\nthat it is present. The four-part records live in\n[reference\u002Fnative-surfaces.md](reference\u002Fnative-surfaces.md).\n\n## Spoke paths\n\nThe `reference\u002F` files write this skill's directory as `\u003Cskill-dir>`, which is\n`${CLAUDE_SKILL_DIR}`, and the plugin's root directory as `\u003Cplugin-root>`, which is\n`${CLAUDE_PLUGIN_ROOT}`. Put each path in place of its placeholder before running a command or\nwriting it into a brief. Those files arrive through the Read tool as plain bytes, so a `${…}` token\nin them would reach the Bash tool unsubstituted, and the Bash tool's environment has no\n`CLAUDE_SKILL_DIR` or `CLAUDE_PLUGIN_ROOT` to expand it from. Basis: the plugins reference,\n\u003Chttps:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Fplugins\u002Fmanifest-reference#where-each-variable-resolves>.\n\n## Related\n\n- `\u002Freview:explain-change` writes a pull request digest; it never posts to the PR.\n- `\u002Fexplainer-video:produce` renders a storyboard into the narrated video, contact sheet and\n  `manifest.json` that the `media` action places.\n- `\u002Freview:reviewer-media` decides which media a PR needs, builds the storyboard from the digest\n  and hands the manifest to the `media` action.\n- `\u002Fplaywright:demo-video` records a live UI flow as PR evidence; the `evidence` action posts it.\n- `\u002Feducation:illustrate` explains a concept, with a video view through the same producer.\n\n## Next\n\n- PR merged, optional review of how the session went: \u002Fsession-flow:retro.\n- `reassess` returned `rework`: \u002Fimplementation:implement.\n- A reviewer walkthrough still to build for the `media` action: \u002Freview:reviewer-media \u003Cpr>.\n\n## Gotchas\n\n- **Verify every agent review finding before presenting it.** Automated reviewers produce incorrect findings often enough that an unverified finding is not evidence; check each against current docs and actual code (step 1.3) before it reaches the user\n- **Never guess at CI failure causes.** Use monitor.md §3.2's prioritized fetch chain: annotations → full log archive → the CLI's failed-log view as last resort. The configured CI-log size cap is `${user_config.fetch_logs_max_bytes}`; when that value is a number other than the 52428800 default (not empty, not a literal unexpanded token), pass it to the archive fetch as `--max-bytes \u003Cvalue>`. The CLI's failed-log view truncates large logs (provider reference \"Checks and CI logs\"); the script-based paths return complete data. Do NOT use broad keyword grep (`error|fail|...`): it matches cleanup steps, variable names, and incidental output\n- **OIDC-based workflows fail when the PR modifies the workflow file.** A forge limitation (provider reference \"Checks and CI logs\"); classify as informational when it applies\n- **Resolving the PR without a number is fragile.** Resolve `\u003Cpr_number>` once at phase entry and pass it to every later forge call (\"PR identity resolution\" above)\n- **Squash merge needs `git branch -D`, not `-d`.** After a squash merge the local branch commit is not in the default branch's history (different SHA), so `-d` says \"not fully merged.\" `-D` is safe because you already confirmed the merge\n- **`git add -A` and `git add .` are banned.** They risk committing secrets, build artifacts, or unrelated changes. Always `git add \u003Cspecific-files>`\n- **Bot comments need reactions AND replies.** React (👍\u002F👎\u002F👀) on every substantive comment AND reply with a per-finding classification table with evidence, for every reviewer. Verify BOTH the reaction and the reply landed on the forge by reading them back. Resolve the thread only when D7.5's conditions are met for EVERY finding in it, and only IF bot-authored; never human-authored, never your own. D7.5 in [reference\u002Fmonitor.md](reference\u002Fmonitor.md)'s per-iteration checklist is the single statement of which dispositions qualify and who may act on one, do not restate it here\n- **Monitor is async, not serial.** Process comments as they arrive while CI is still running; bots post at different times\n- **Check mergeable BEFORE polling CI.** While the PR's mergeability reports a conflict, the forge runs no workflows. Integrate the default branch first, then poll\n- **A green review lane can mean nothing was reviewed.** These lanes report on their session, not their output: a session that ends without error goes green even when it posted no review and no finding. No guard beside the lane judges what it posted; B2 ([reference\u002Fmonitor.md](reference\u002Fmonitor.md)'s per-iteration checklist) is the only productivity check, and billing is no signal, since the session is billed either way. Judge a review lane by what it produced (B2), never by its check row, and run the review locally when it produced nothing\n- **Never merge with unclassified FAILURE check runs.** Every FAILURE must be investigated, classified (real failure vs informational), and documented before merge is even suggested. See [readiness.md](reference\u002Freadiness.md) for the full 6-gate checklist\n- **\"No comments\" does NOT mean \"ready.\"** Comment-only actors post at unpredictable times; a 2-minute cooldown after the last check-run completion or comment arrival prevents the race. See readiness.md Gate 5\n- **Security scans are always blocking.** Any check run or bot comment reporting security findings (secrets, vulnerabilities) triggers mandatory triage; even a false positive must be explicitly classified and documented before merge\n- **Discover actors, don't hardcode them.** Security tools and AI reviewers change. Monitor discovers actors from the PR's checks and comments, classifies them by category (CI, security, review), and evaluates accordingly\n- **Uncommitted changes are silently lost on branch deletion.** `git reflog` recovers only commits. Before staging (Phase 2.3.1) and before the post-merge worktree transition (Phase 4.3), check `git status --porcelain` for unrelated uncommitted changes and stash them (`git stash push -m \"desc\" -- \u003Cfiles>`); stashes survive branch deletion. Never silently ignore uncommitted changes\n- **Cloud sessions poll.** Autonomous cloud sessions (`CLAUDE_CODE_REMOTE=true`) poll the PR's checks and comment surfaces on a fixed 60-90s cadence, and subscribe to PR activity where offered so reviews and failed checks wake them sooner. The poll stays, since pushes and green checks produce no event (§3.0.0 of monitor.md holds the decision, its pointer and its recheck trigger)\n- **Monitor checks the push channel first, then falls back.** Three tiers on local CLI sessions: (1) **push channel** (when your environment ships a forge-events MCP channel; ~0 idle requests), (2) **Monitor tool** (a `Monitor` watch armed with an explicit `timeout_ms` and re-armed at each deadline, per monitor.md §3.0.1; 60s poll fires on real CI\u002Fcomment events; cancel via `TaskStop`), (3) fixed-interval cron polling (deprecated, wasteful). Check for a channel before the Monitor tool: polling spends ~1 request per 30s interval vs ~0 idle with push delivery\n","**Arguments.** `[prep|create|ready|monitor|comments|ci|merge|reassess|evidence|status|full|…] [--unattended] [args]`. Full set: the Arguments table below. e.g., \u002Fpull-request prep, \u002Fpull-request create, \u002Fpull-request ready, \u002Fpull-request monitor, \u002Fpull-request comments, \u002Fpull-request merge, \u002Fpull-request evidence \u003Cpr> --evidence \u003Cfile>, \u002Fpull-request media \u003Cpr> --manifest \u003Cfile>, \u002Fpull-request status, \u002Fpull-request full, \u002Fpull-request fetch-logs \u003Cpr|run>, \u002Fpull-request view \u003Cpr> --diff, \u002Fpull-request list --head \u003Cbranch>, \u002Fpull-request review-venue\n\n## Repository context. Gather first\n\nThe read actions `view`, `list`, `review-comments` and `review-venue`, `comments --ci`, `conflicts --ci`, `dependency --ci` and `ci --ci` skip this section. For every other action, collect these with\n**individual** Bash calls, one command per call, never combined into a single invocation:\n\n- Current branch, `git branch --show-current`\n- Recent commits, `git log --oneline -5`\n- Working tree status, `git status --porcelain`\n- Changed files (staged+unstaged), `git diff --name-only HEAD`\n\nTreat a failure (not a repository, git unavailable) as an unknown value and carry on. Keep them\nseparate body calls, not pre-computed, so the skill stays invocable from inside a worktree: the composition record is the [gather-block convention](https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fdocs\u002Fconventions\u002Fgather-block\u002FREADME.md#the-pre-compute-block-runs-as-one-shell-invocation).\n**Pointer**: when an isolated session refuses a git command, fetch [How Claude Code enforces isolation](https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Fworktrees#how-claude-code-enforces-isolation) live.\n\n## Purpose\n\nRun the PR lifecycle from quality review through merge and the post-merge worktree transition, detecting state and resuming.\n\n**The two non-negotiable gates:**\n\n1. **Finding verification** (prep phase): agent review findings have a demonstrated error rate, so every finding is verified against current docs and actual code before it reaches the user.\n2. **Research-gated CI fixes** (monitor phase): no code fix without multi-source research consensus on the root cause. Unresearched \"obvious\" CI fixes are how wrong fixes ship.\n\n## Adapting to your environment (graceful degrade)\n\nThis skill is self-contained: it runs on `git`, the forge's CLI, `jq`, and its bundled scripts (skill-private ones under `${CLAUDE_PLUGIN_ROOT}\u002Fskills\u002Fpull-request\u002Fscripts\u002F`, plugin-shared ones under `${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002F`, which the `reference\u002F` files call `\u003Cscripts-dir>`: substitute the resolved path before a command from them reaches Bash). `jq` is required (the merge and readiness paths pipe forge API output through it): check `command -v jq` before the first phase that parses; if missing, stop with the install remediation (\u003Chttps:\u002F\u002Fjqlang.org\u002Fdownload\u002F>; a separate install under Git Bash on native Windows) instead of failing mid-phase. The readiness thread gate (Gate 7) also needs Python 3.11+; without it that gate holds as unproven and the rest of the skill runs. The `media` action and the stale reviewer-media check need Python 3.11+ too: without it `media` stops with the install remediation (\u003Chttps:\u002F\u002Fwww.python.org\u002Fdownloads\u002F>) and the stale check is skipped. An adjacent capability a phase names (a code-review skill or agents, a simplifier, a build\u002Ftest\u002Flint verifier, an external research skill, an exploration skill, a work-item tracker, a CI-log-audit agent, a forge-events push channel) is **optional**: invoke it when your environment provides it (a skill, plugin, agent, or MCP server), else follow the inline guidance, which stands on its own. Never block a phase because an adjacent tool is absent.\n\nConsumer conventions come from the consuming project's own `CLAUDE.md`, `AGENTS.md`, and rules. Notably: PR body template, branch naming, merge style (this skill defaults to squash), review-reply identity (some projects post bot-identity replies via a wrapper; default is the forge CLI under your own login), and any extra pre-PR gates. Read them before creating or merging.\n\n**Provider commands.** This body names forge operations (resolve the PR, read its checks, reply on a thread); their GitHub commands (the one forge supported today) are in [the GitHub provider reference](..\u002F..\u002Freference\u002Fproviders\u002Fgithub\u002FREADME.md). Read it before the first forge operation. The phase files under `reference\u002F` carry each phase's full GitHub procedure.\n\n**PR title format** resolves via the same ladder `\u002Fsource-control:commit` uses for the commit subject (see its SKILL.md), checked in order: the `pr_title_pattern` resolved across the layers of the source-control convention config (team value in `docs\u002Fconventions\u002Fsource-control.yaml`) per [..\u002F..\u002Freference\u002Fconfig-resolution.md](..\u002F..\u002Freference\u002Fconfig-resolution.md) → the consuming project's own `CLAUDE.md`\u002F`AGENTS.md`\u002Frules → Conventional Commits (11-type vocabulary. `build, chore, ci, docs, feat, fix, perf, refactor, revert, style, test`; `security` is not a member. **Pointer**: when the default set is questioned, fetch commitlint's [config-conventional `type-enum`](https:\u002F\u002Fgithub.com\u002Fconventional-changelog\u002Fcommitlint\u002Ftree\u002Fmaster\u002F%40commitlint\u002Fconfig-conventional#type-enum) live. **As of**: 2026-11-10. **Recheck trigger**: that rule's value changes) as the default. See [reference\u002Fcreate.md](reference\u002Fcreate.md) §2.4.1 for where the title is derived. When no layer resolves a `pr_title_pattern` and nothing is inferable, point the user at `\u002Fsource-control:setup`.\n\n**PR-body attribution** (the `🤖 Generated with [Claude Code]…` line) resolves from the `pr_body_attribution` key across the same layers (team YAML first, [..\u002F..\u002Freference\u002Fconfig-resolution.md](..\u002F..\u002Freference\u002Fconfig-resolution.md)). Absent → the default line, `none` → omitted, any other value → that line. It is the PR-body analogue of `\u002Fsource-control:commit`'s `trailer_policy` and gated separately; see [reference\u002Fcreate.md](reference\u002Fcreate.md) §2.4.1 for the assembly.\n\n**Branch-to-issue grammar** is a POSIX ERE whose LAST capture group is the work-item reference (an issue number or a tracker key), e.g. `^[^\u002F]+\u002F([0-9]+)-`. It resolves first from the `branch_issue_pattern` key across the layers (team value in the YAML; [..\u002F..\u002Freference\u002Fconfig-resolution.md](..\u002F..\u002Freference\u002Fconfig-resolution.md)), which `parse-branch-issue.sh` reads itself. The `branch_issue_pattern` userConfig is a deprecated fallback, read only when no layer sets the key: its value is `${user_config.branch_issue_pattern}` (this line is the substituted surface; a literal `${user_config.branch_issue_pattern}` token means unset), and [reference\u002Fcreate.md](reference\u002Fcreate.md) §2.4.0 fills its `\u003Cbranch-issue-pattern>` slot with a real ERE from it. When the script falls back to that value it prints a deprecation note on stderr; relay it to the user. Neither set: a bound work-item tracker's adapter supplies the grammar (§2.4.0), else the built-in `\u003Ctype>\u002F\u003CN>-\u003Cslug>` (and `routine-issue-\u003CN>`) convention.\n\n**PR open state** (`ready` or `draft`) decides whether `create` takes the new PR through `ready` or leaves it a draft. A draft means a person asked to see the work early or the PR is a multi-session container; ready means complete and through `ready`'s gates. The per-user value is `${user_config.pr_open_state}` (this line carries the substituted value; a literal `${user_config.pr_open_state}` token means unset). A `pr_open_state` key in the repository's `docs\u002Fconventions\u002Fsource-control.yaml`, read at the PR's base branch and never at its head, wins over it; a request for a draft wins over both; with none set the state is `ready`, and a consumer opts into `draft`. An unattended run follows the resolved value like an attended one, and `ready --unattended`'s holds still apply. [reference\u002Fcreate.md](reference\u002Fcreate.md) §2.4.3 holds the state meanings, resolves the state and reports the supplying level; the levels and the invalid-value rule are in [..\u002F..\u002Freference\u002Fconfig-resolution.md](..\u002F..\u002Freference\u002Fconfig-resolution.md).\n\n**Review venue** (`auto`, `local`, `ci` or `both`) decides which local review steps run before and at the ready flip. The per-user value is `${user_config.review_venue}` (this line carries the substituted value; a literal `${user_config.review_venue}` token means unset), passed as `--user` to `scripts\u002Fread-review-venue.sh`. A `review_venue` key in the repository's `docs\u002Fconventions\u002Fsource-control.yaml`, read at the PR's base branch, wins over it. [reference\u002Freview-venue.md](reference\u002Freview-venue.md) reads it and owns what each value does.\n\n**PR body preview** (`post` or `preview`) decides whether an attended run shows a body and waits for a yes before posting it. The per-user value is `${user_config.pr_body_preview}`; **PR body refresh** (`true` or `false`) decides whether a push to an open PR also rewrites the body sections it made stale, per-user value `${user_config.pr_body_refresh}`. A literal unexpanded token means unset. A key of the same name in the repository's `docs\u002Fconventions\u002Fsource-control.yaml` wins over each; the levels are in [..\u002F..\u002Freference\u002Fconfig-resolution.md](..\u002F..\u002Freference\u002Fconfig-resolution.md). [reference\u002Fcreate.md](reference\u002Fcreate.md) §2.4.3 applies the preview and [reference\u002Fmonitor.md](reference\u002Fmonitor.md) §3.3.2 the refresh.\n\n## Emit checklist\n\nFor runs spanning 3+ phases, copy `${CLAUDE_PLUGIN_ROOT}\u002Fskills\u002Fpull-request\u002Ftemplates\u002Fchecklist.md` into your project's working-notes location (or track it inline) and tick each `- [ ]` as its phase produces output. It survives `\u002Fclear`.\n\n## Arguments\n\n`$ARGUMENTS`. Action selector:\n\n| Action | Entry point | Use case |\n|--------|-------------|----------|\n| *(empty)* | Smart default | Detect current state, resume from right phase |\n| `prep` | Phase 1 | Review (skipped when the CI code-review lane covers it, [reference\u002Freview-venue.md](reference\u002Freview-venue.md)) + verify findings + simplify + comment pass + verify build |\n| `prep quick` | Phase 1 (fast) | Code errors only, skip simplify and the comment pass |\n| `prep review-only` | Phase 1 (partial) | Just review + verify findings |\n| `prep simplify-only` | Phase 1 (partial) | Just simplify + comment pass + re-verify |\n| `create [--draft '\u003Creason>']` | Phase 2 | Branch-name check + commit + changelog fragments where the repo releases from them ([reference\u002Fcreate.md](reference\u002Fcreate.md) §2.3.3) + push + open the PR as a draft, its body carrying the plan Brief's acceptance criteria when the branch implements a plan and `Closes` only for an issue whose done-when items the diff delivers, else `Refs:` with a `## Remaining` section ([reference\u002Fcreate.md](reference\u002Fcreate.md) §2.4.0, §2.4.1), then run `ready` on it when the resolved PR open state is `ready`; `--draft` requests a draft whose `Draft:` line names the reason. Reports the PR URL, its open state and the level that set it, then arms event delivery ([reference\u002Fcreate.md](reference\u002Fcreate.md) §2.6) when the session keeps acting on the PR, else stops |\n| `create --pushed --worktree \u003Cpath> [--draft '\u003Creason>'] [--hold '\u003Cline>']` | Phase 2 (PR-only) | **PR-only entry for an orchestrated flow**, the branch is already committed and pushed (by a dispatched worker), so this skips commit \u002F push \u002F branch update, re-resolves branch and diff from the given target worktree (not the session cwd), and runs body assembly + gates + opens the PR for `\u003Cbranch>`, then follows the resolved open state like an attended run (the repository level is read at the base branch, never from the worktree). `--draft` requests a draft whose `Draft:` line names the reason; `--hold` puts a `Hold:` line in the body the PR opens with. Used by `\u002Fwork-items:work`'s orchestrator after its pre-PR gate. See [reference\u002Fcreate.md](reference\u002Fcreate.md) §2.7 |\n| `ready` | Phase 2.5 | Update the branch from its base by the method the shared branch-update decision chooses (`ready --unattended` never rewrites a pushed head), run the code and security reviews over the PR diff in fresh-context subagents (each skipped when its CI review lane is present and healthy, [reference\u002Freview-venue.md](reference\u002Freview-venue.md)) and the verify gate on the updated head (its test suites skipped when CI's test lanes already passed on that head, reported as `skipped (CI covers it)`; `ready --verify-local` runs them anyway), flip the draft to ready (the base branch's `pre_ready` list can turn off any of its six configurable steps; each step's outcome is reported), post the `evidence` comment when an evidence file exists for the confirmed head (`--evidence \u003Cfile>` names one; attended runs only, and a refusal is reported, never undoes the flip), then hand the PR to `\u002Freview:explain-change`, whose `digest_policy` decides whether a digest is offered or built, then to `\u002Freview:reviewer-media`, whose policy decides whether a narrated walkthrough is built (neither gates anything). Never arms auto-merge: ready starts the AI reviews. Refuses on a branch with no PR, and leaves in draft a held or superseded draft, or one whose body closes an issue the diff does not finish. `ready --unattended` (every lane, loop or dispatched caller) also leaves a draft a person asked for (its `Draft:` body line) in draft. `ready [\u003Cpr>] --skip \u003Cstep>[,\u003Cstep>...] --reason \"\u003Ctext>\"` lets a person present skip configurable steps for this run only, reported and recorded in one PR comment, never in the body; refused unattended. See [reference\u002Fready-for-review.md](reference\u002Fready-for-review.md) |\n| `monitor` | Phase 3 | Watch CI, fix failures, evaluate comments. **Three-tier event delivery: (1) push channel** when your environment ships a forge-events channel (an MCP server delivering webhook events into the session), ~0 idle requests; **(2) Monitor tool** fallback (60s poll); **(3) polling** in cloud\u002Fheadless sessions, plus the PR-activity subscription where the session offers it. Check the push channel FIRST per [monitor.md](reference\u002Fmonitor.md) §3.0.05 before falling back |\n| `comments` | Phase 3.5 | Evaluate\u002Frespond to PR comments only |\n| `comments --ci` | Phase 3.5 (CI) | The pr-unblock lane's unattended mode: the runner's kept reviewer-bot threads, or an earlier pr-verify FAIL's notes, only, through its review-thread tool, running no head code. See [reference\u002Fcomments-ci.md](reference\u002Fcomments-ci.md) |\n| `conflicts --ci` | Phase 2.5.2 (CI) | Unattended conflict resolution for the pr-unblock lane: reads the runner's `conflicts.json` as data, resolves each conflicted path in place under the base-merge step's discipline, runs no command, never edits a path under `.github\u002Fworkflows\u002F`, and leaves a path whose two intents cannot both hold with its markers, which escalates it. See [reference\u002Fconflicts-ci.md](reference\u002Fconflicts-ci.md) |\n| `dependency --ci` | Phase 3.5 (CI) | The pr-unblock lane's unattended review of one risky Dependabot update: reads the context file, the checkout and the upstream release notes as data, leaves a scoped fix outside `.github\u002F` uncommitted for a later job to push, and returns a merge, fix, close or escalate verdict as structured output, running no head code. See [reference\u002Fdependency-ci.md](reference\u002Fdependency-ci.md) |\n| `ci --ci` | CI | The pr-unblock lane's unattended fix of a PR's failing checks: reads the runner's context file as data, reproduces each check with the repository's own command, fixes the cause, reruns it, and leaves the fix uncommitted for the runner's patch; never weakens a check, and escalates what it cannot fix with a named reason. See [reference\u002Fci-ci.md](reference\u002Fci-ci.md) |\n| `merge` | Phase 4 | Squash merge, or merge or enqueue without asking once the PR is `CLEAN` and the AI review checks have passed on the reviewed head, with the plain merge command (a refusal is reported, never handed to the user or retried in another shape), then confirm a merge-queue PR entered the queue (enqueue a `CLEAN` one left out, report the refusal text verbatim; §4.2.2) and wait for `MERGED` (no strategy flag behind a merge queue, `--auto` only pinned to the reviewed head; [reference\u002Fmerge.md](reference\u002Fmerge.md) §4.2.1) + worktree transition (reuse it, or release this session's lock) + verify. Never removes a worktree: that is `\u002Fsource-control:worktree cleanup` |\n| `evidence \u003Cpr> --evidence \u003Cfile> [--attach \u003Cpath>]... [--summary \u003Cfile>] [--artifact-url \u003Curl> [--retention-days \u003Cn>]]` | Evidence | After the verifier passed for the PR's head, post one evidence comment, or update it in place for a new head, with `gh pr comment --edit-last --create-if-none --attach`: the text summary, publishable captures only in collapsed sections laid out by `templates\u002Fevidence-comment.md`, and the CI capture artifact link with its expiry. A file outside the `publishable` list, or a leak-scan hit on the body, posts nothing. See [reference\u002Fevidence.md](reference\u002Fevidence.md) |\n| `media \u003Cpr> --manifest \u003Cfile> [--artifact-url \u003Curl> [--retention-days \u003Cn>]]` | Reviewer media | Place a produced walkthrough (video, contact sheet, chapter list, screenshots) in the marked `## Reviewer media` section near the top of the PR body, rewriting only the text between its two marker lines: claim the head first, attach only the files the producer's `manifest.json` lists with `gh pr edit --attach`, else link the Actions artifact with its expiry, and swap the request label for the posted one. A leak-scan hit, a screenshot with no frame-text sidecar, a claim another run holds, a body that changed under the edit, or no way to host the media posts nothing. `media stale \u003Cpr>` marks a section that shows an older commit; `media shows --facts \u003Cfile>` prints the commit it shows; `media fail \u003Cpr>` records a failed build once per commit. See [reference\u002Fmedia.md](reference\u002Fmedia.md) |\n| `reassess [\u003Cpr>] [--act] [--state-dir \u003Cdir>]` | Reassess | Judge whether one open PR should still land, even when green and mergeable, and return `continue`, `rework` or `close` with its evidence: report-only unless `--act` carries out the verdict. Runs on one PR and is no merge prerequisite. `--state-dir` defaults to `${CLAUDE_PLUGIN_DATA}\u002Fstate\u002Fbabysit-prs`. See [reference\u002Freassess.md](reference\u002Freassess.md) |\n| `status` | Report only | Unified status across all phases |\n| `full` | Phase 1-4 | Run prep → create → monitor → merge end-to-end |\n| `fetch-logs \u003Cpr\\|run> [--raw\\|--job \u003Cjob-id>]` | CI log retrieval | Pull failed-CI evidence: default = `::error`\u002F`::warning` annotations only (cheapest); `--raw` = full ZIP dump for archive review; `--job \u003Cid>` = per-job plain text |\n| `view [\u003Cpr>] [--repo \u003Cowner\u002Frepo>] [--diff] [--out \u003Cfile>]` | Read only | One PR's facts as JSON, or its diff with `--diff`; no number reads the current branch's PR. See [reference\u002Fread.md](reference\u002Fread.md) |\n| `list [--head \u003Cbranch>\\|--head-match \u003CERE>] [--search \u003Ctext>] [--state open\\|closed\\|merged\\|all] [--repo \u003Cowner\u002Frepo>]` | Read only | PRs as a JSON array, open by default |\n| `review-comments (--prs \u003Cn> \\| --pr \u003Cnumber>) [--repo \u003Cowner\u002Frepo>]` | Read only | The review comments of the last `\u003Cn>` merged PRs, or of one PR, as JSON lines. See [reference\u002Fread.md](reference\u002Fread.md) |\n| `review-venue [--repo \u003Cowner\u002Frepo>] [--max-age-days \u003Cn>] [--head \u003Csha>]` | Read only | Per AI review lane (code, security): `ci` when the repository's CI lane is present and healthy, so no local review runs for it, else `local`; with `--head`, a `tests` line: `ci` when CI's test lanes passed on that commit. See [reference\u002Freview-venue.md](reference\u002Freview-venue.md) |\n\nFor the all-PR continuous loop (discover every open PR, work each to readiness, self-pace), use\nthe sibling skill `\u002Fsource-control:babysit-prs`. It wraps this skill's per-PR review discipline\nin fleet orchestration and never merges.\n\n## Action defaults\n\n- **Merge mode:** `merge` squash-merges (one squashed commit per PR onto the default branch), see [reference\u002Fmerge.md](reference\u002Fmerge.md) §4.2. Follow the consuming project's convention when it differs. A stacked PR lands every layer below it: see [stacked-prs.md](..\u002F..\u002Freference\u002Fstacked-prs.md).\n- **Stack or separate:** before `create` opens a PR that builds on another open PR, decide stack versus sequential PRs off the trunk by \"When to stack\" in [stacked-prs.md](..\u002F..\u002Freference\u002Fstacked-prs.md).\n- **Monitor cadence:** `monitor` polls the PR's checks (the §3.0.1 REST read) and comment fetches every 60 seconds, see [reference\u002Fmonitor.md](reference\u002Fmonitor.md) §3.1; every loop follows the [API budget](..\u002F..\u002Freference\u002Fproviders\u002Fgithub\u002FREADME.md#api-budget). A wait is a REST poll that runs in the background (the §3.0.1 Monitor watch, or a `run_in_background` poll where Monitor is unavailable), never a foreground `sleep` or `until` loop and never the CLI's blocking checks watch, and it first reports each pending job as queued or running (\"Waiting on a pending check\"). A dispatched subagent does not wait: it stops once its PR is marked ready and reports the full head sha, and the dispatching session watches and merges ([reference\u002Fmonitor.md](reference\u002Fmonitor.md)).\n- **Required reviewers:** `create` requests no reviewers when it opens the PR. See [reference\u002Fcreate.md](reference\u002Fcreate.md) §2.4.3.\n\n## PR identity resolution\n\nPR identity (number, URL) resolves **live from the forge**, the authoritative source, at every phase that needs it. No caching, no state files.\n\n**Standard pattern** (every phase): resolve the current branch's PR number; when that is ambiguous (multiple open PRs, stale checkout, returning days later), name the branch explicitly. Both commands are in the provider reference's \"Resolve a pull request\".\n\nAfter resolving once at phase entry, **pass `\u003Cpr_number>` explicitly to all subsequent forge calls** in that phase (reading checks, merging, etc.). Never rely on bare-branch resolution mid-phase.\n\n**Why explicit PR numbers?** Resolution with no number goes by the HEAD branch, which fails when the worktree was cleaned up (branch context lost), multiple open PRs exist (wrong match), you return days later (stale checkout), or another session's PR merged first. Resolving once per phase bounds this risk to one point.\n\n---\n\n## Phase 0: Parse action and detect state\n\nParse `$ARGUMENTS` to extract the action (first token) and any sub-arguments, after taking out\n`--unattended` wherever it appears: it is the caller declaring that nobody is present. Pass it on\nto `ready` (which takes it as its own flag) and to no script; with nothing else left, the action\nis the smart default.\n\n**Smart default** (empty args): read live state from the forge to determine the right phase:\n\n```text\n1. Check git branch — on the default branch? → \"Create a worktree or branch first\"\n2. Resolve PR for current branch, reading its number, state and draft flag:\n   a. no PR found → no PR yet → START AT PHASE 1 (prep); skip steps 3-6\n      entirely (they all need a PR number that does not exist yet)\n   b. state = MERGED → skip to Phase 4.3 (worktree transition only: new branch from the default branch, delete the merged local branch)\n   c. state = CLOSED → report \"PR was closed without merging\" and stop\n   d. state = OPEN → capture pr_number, continue to step 3\n3. isDraft = true → START AT PHASE 2.5 (ready); skip steps 4-6. A draft owes its\n   review and its flip, not monitoring: no reviewer has been asked to read it yet\n4. Read the PR's checks — still running? → start at monitor\n5. Check for unaddressed comments → start at monitor (comments sub-phase)\n6. CI green + comments addressed → suggest merge\n```\n\nPresent the detected state and proceed to its phase, announcing it in interactive mode; in autonomous mode (`full`), proceed without pausing. Phase transitions are not decision points.\n\n**Status action**: read the PR's number, URL and state, then report current state across all phases.\n\n---\n\n## Phases\n\nExecute in order. Each phase is self-contained. Read the relevant file for detailed steps:\n\n| Phase | File | Entry actions |\n|-------|------|--------------|\n| 1. Prep | [reference\u002Fprep.md](reference\u002Fprep.md) | `prep`, `prep quick`, `prep review-only`, `prep simplify-only` |\n| 2. Create | [reference\u002Fcreate.md](reference\u002Fcreate.md) | `create` |\n| 2.5. Ready for review | [reference\u002Fready-for-review.md](reference\u002Fready-for-review.md) | `ready` |\n| 3. Monitor | [reference\u002Fmonitor.md](reference\u002Fmonitor.md) | `monitor`, `comments` |\n| 2.5.2 (CI) | [reference\u002Fconflicts-ci.md](reference\u002Fconflicts-ci.md) | `conflicts --ci` |\n| 3.5 (CI) | [reference\u002Fcomments-ci.md](reference\u002Fcomments-ci.md) | `comments --ci` |\n| 3.5 (CI) | [reference\u002Fdependency-ci.md](reference\u002Fdependency-ci.md) | `dependency --ci` |\n| CI | [reference\u002Fci-ci.md](reference\u002Fci-ci.md) | `ci --ci` |\n| 4. Merge | [reference\u002Fmerge.md](reference\u002Fmerge.md) | `merge` |\n| Evidence | [reference\u002Fevidence.md](reference\u002Fevidence.md) | `evidence` |\n| Reviewer media | [reference\u002Fmedia.md](reference\u002Fmedia.md) | `media` |\n| Reassess | [reference\u002Freassess.md](reference\u002Freassess.md) | `reassess` |\n\n---\n\n## Monitor entry\n\nOn `monitor`, `comments`, or `full` reaching monitor, read [reference\u002Fmonitor.md](reference\u002Fmonitor.md) before any CI polling or comment evaluation: its entry checklist runs first, and its per-iteration checklist governs every CI or comment event. Standalone `create` runs Steps 1-3 of that entry checklist when its session will keep acting on the PR it opened ([reference\u002Fcreate.md](reference\u002Fcreate.md) §2.6); `create --pushed` leaves them to its caller. `comments --ci` skips it, and `conflicts --ci`, `dependency --ci` and `ci --ci` do not enter Phase 3.\n\n---\n\n## Full lifecycle (`\u002Fsource-control:pull-request full`)\n\n`full` chains prep, create, monitor, and merge in one invocation. Read\n[reference\u002Ffull-lifecycle.md](reference\u002Ffull-lifecycle.md) when invoked with `full`, and only\nthen: it owns the phase-to-phase handoff state, the abort points, and what `full` does\ndifferently from running the four phases by hand. Any other action routes through the phase table\nabove instead.\n\n## Fetch CI logs (`\u002Fsource-control:pull-request fetch-logs \u003Cpr|run> [--raw|--job \u003Cjob-id>]`)\n\nPublic action for failed-CI evidence, a tiered fetch chain: cheapest signal first, escalating only when the lower tier is insufficient. Consumers say WHAT they want; the skill picks the internal helper from the flags.\n\n**Behaviors:**\n\n| Invocation | What it returns | When to use |\n|------------|-----------------|-------------|\n| `fetch-logs \u003Cpr-number>` (default) | `::error::` + `::warning::` annotations across all failed jobs | First-pass. Usually enough to identify the cause |\n| `fetch-logs \u003Cpr-number> --failed` | Annotations from failed jobs only | When the run has many jobs and noise is a concern |\n| `fetch-logs \u003Crun-id> --raw` | Full CI log archive, dumped in scope | When annotations are sparse \u002F missing. Debug-grade detail |\n| `fetch-logs \u003Crun-id> --job \u003Cjob-id>` | Plain-text log of one job | Targeted dive after seeing which job failed |\n\n`\u003Cpr-number>` and `\u003Crun-id>` are interchangeable; given a PR number, the skill resolves its latest run.\n\n**Composition with `monitor`:** `monitor` invokes this action on CI failure. Direct `fetch-logs` is for ad-hoc post-mortem (e.g., a closed PR's CI failure, auditing a green run for warnings).\n\n**Implementation note:** this action delegates to the bundled, private `fetch-annotations.sh` and `fetch-failed-logs.sh` scripts. Consumers MUST NOT cite script paths; use this action.\n\n## Read a pull request (`view`, `list`)\n\n`view` and `list` are the read-only actions other skills call instead of a forge CLI. Read\n[reference\u002Fread.md](reference\u002Fread.md) when invoked with either, and only then: it owns the\ncommand, the output fields, and the exit codes.\n\n## Read the review venue (`review-venue`)\n\n`review-venue` is the read action prep, ready and other skills call before a local AI review, so a\nreview runs in CI or locally, never in both. Read\n[reference\u002Freview-venue.md](reference\u002Freview-venue.md) when invoked with it, or when a phase\nreaches a review: it owns the command, what makes a lane healthy, and how each line is applied and\nreported.\n\n---\n\n## Important notes\n\n- **Side effects**: this skill commits, pushes, creates PRs, and merges. User approval gates at each dangerous step (commit message, CI fix, merge outside the [reference\u002Fmerge.md](reference\u002Fmerge.md) §4.2.1 merge after AI review) are the human checkpoints the skill enforces\n- **Finding verification is non-negotiable** (the first gate under Purpose)\n- **Research-driven fixes** are the point of the monitor phase: a short research burst costs little; an unresearched fix costs a lot\n- **Max 3 CI fix iterations**, to stop fix-push-fail loops\n- **Findings triage**:\n  - **Bot comments classified CORRECT** (Codex, claude-review, etc., after evidence-based verification against actual code): **auto-fix + test + react 👍 + D5 reply in the same turn, then push with the round's other fixes and post the D7 reply** ([monitor.md](reference\u002Fmonitor.md) §3.3.2). No user-approval pause: the CORRECT\u002FINCORRECT classification is the safety gate. Applies when the fix is small and scoped (\u003C~50 LOC, single concern); pause for cross-cutting refactors even when CORRECT\n  - **Bot comments classified INCORRECT**: autonomous 👎 reaction + reply with research-backed counter-evidence. Never silently ignore\n  - **Human reviewer comments**: always pause for user approval before reacting or fixing, regardless of classification\n- **Prep skips the review\u002Fsimplify work when no `code` class is detected**, and the verify gate reduces to lint. `ready` still reviews a diff that touches an instruction surface ([definition](reference\u002Fready-for-review.md#docs-only-and-instruction-surfaces)). Extra project-specific prep-evidence requirements come from the consuming project's own hooks\n\n---\n\n## Boundary, native Claude Code surfaces\n\nThree native surfaces cover parts of this lifecycle and get conflated with it when a PR is opened\nor watched.\n\n- **`pr` (bundled skill)**: creates one pull request generically; it gathers branch\n  context and applies Claude Code's own title, body, and attribution. The model and\n  the person can both invoke it.\n- **`\u002Fcommit-push-pr` (built-in command)**: commits, pushes, and opens a PR in one prompt-driven\n  step. The model and the person can both invoke it.\n- **`\u002Fautofix-pr` (built-in command)**: spawns a cloud session that watches the current branch's\n  PR and pushes fixes when CI fails or reviewers comment. Reserved for the\n  person to run; the model does not invoke it.\n- **This skill (marketplace plugin).** The whole lifecycle: prep with verified findings, a draft\n  under this repository's PR title and body contract, the ready flip after updating the branch from its base,\n  research-gated local monitoring with per-finding classification and replies, and merge.\n\n**Routing.** Where no draft discipline, body contract, or later lifecycle step applies: when the\nbundled `pr` skill resolves in this session, prefer it to open a PR from work already committed;\nwhen `\u002Fcommit-push-pr` resolves, prefer it only when the whole working tree belongs in the commit,\nsince it commits, pushes and opens the PR in one step. With unrelated uncommitted changes, `pr`\nis the native route, never `\u002Fcommit-push-pr`. Prefer\nthis skill whenever the repository declares a PR convention or the work continues into ready,\nmonitor, or merge.\n\n**Choose who watches the PR.** At the end of `ready`, and at `monitor` entry when no choice was\nmade, pick between this skill's monitor, a background agent, `\u002Fbackground`, `\u002Fautofix-pr` and the\nbabysit loop by whether the person is staying and whether the machine stays on, after reading the\nplan's rate-limit windows: at 95% or above, report the window and the options, and the person decides. For a route only\nthe person can start, print it ready to run. If \u002Fautofix-pr is available in your session (gate\nbasis: it registers gated and needs forge CLI and cloud-session access; records in\n[reference\u002Fnative-surfaces.md](reference\u002Fnative-surfaces.md)), the person can run it instead of or\nalongside this skill's monitor, with a filled-in prompt carrying this skill's review discipline. The\nmatrix and the prompt:\n[reference\u002Fwatch-handoff.md](reference\u002Fwatch-handoff.md). When the caller passes `--unattended`,\nwith any action, the run records that recommendation and the `\u002Fautofix-pr` prompt in its output\nand does not ask the person; without that flag it asks.\n\n**Mutation gate.** `pr` and `\u002Fcommit-push-pr` commit, push, and open a PR; `\u002Fautofix-pr` pushes\nfixes to the PR branch from a cloud session. This skill never chains into any of them on its own\nbehalf. When `\u002Fautofix-pr` runs alongside this loop, both push to one branch. Before each fix\ncommit, fetch the PR head and merge it into the checked-out branch (`git merge --ff-only` when there\nare no local commits, else `git merge`), then re-check the pending edit against the merged tree and\ndrop it when the cloud session already fixed that finding. A fetch alone leaves the local branch on\nthe old tip, and the push that follows is rejected as non-fast-forward.\n\n**Availability is never assumed.** `pr` and `\u002Fautofix-pr` register gated, and `\u002Fautofix-pr` also\nneeds forge CLI and cloud-session access; this section states what to do when a surface resolves, never\nthat it is present. The four-part records live in\n[reference\u002Fnative-surfaces.md](reference\u002Fnative-surfaces.md).\n\n## Spoke paths\n\nThe `reference\u002F` files write this skill's directory as `\u003Cskill-dir>`, which is\n`${CLAUDE_SKILL_DIR}`, and the plugin's root directory as `\u003Cplugin-root>`, which is\n`${CLAUDE_PLUGIN_ROOT}`. Put each path in place of its placeholder before running a command or\nwriting it into a brief. Those files arrive through the Read tool as plain bytes, so a `${…}` token\nin them would reach the Bash tool unsubstituted, and the Bash tool's environment has no\n`CLAUDE_SKILL_DIR` or `CLAUDE_PLUGIN_ROOT` to expand it from. Basis: the plugins reference,\n\u003Chttps:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Fplugins\u002Fmanifest-reference#where-each-variable-resolves>.\n\n## Related\n\n- `\u002Freview:explain-change` writes a pull request digest; it never posts to the PR.\n- `\u002Fexplainer-video:produce` renders a storyboard into the narrated video, contact sheet and\n  `manifest.json` that the `media` action places.\n- `\u002Freview:reviewer-media` decides which media a PR needs, builds the storyboard from the digest\n  and hands the manifest to the `media` action.\n- `\u002Fplaywright:demo-video` records a live UI flow as PR evidence; the `evidence` action posts it.\n- `\u002Feducation:illustrate` explains a concept, with a video view through the same producer.\n\n## Next\n\n- PR merged, optional review of how the session went: \u002Fsession-flow:retro.\n- `reassess` returned `rework`: \u002Fimplementation:implement.\n- A reviewer walkthrough still to build for the `media` action: \u002Freview:reviewer-media \u003Cpr>.\n\n## Gotchas\n\n- **Verify every agent review finding before presenting it.** Automated reviewers produce incorrect findings often enough that an unverified finding is not evidence; check each against current docs and actual code (step 1.3) before it reaches the user\n- **Never guess at CI failure causes.** Use monitor.md §3.2's prioritized fetch chain: annotations → full log archive → the CLI's failed-log view as last resort. The configured CI-log size cap is `${user_config.fetch_logs_max_bytes}`; when that value is a number other than the 52428800 default (not empty, not a literal unexpanded token), pass it to the archive fetch as `--max-bytes \u003Cvalue>`. The CLI's failed-log view truncates large logs (provider reference \"Checks and CI logs\"); the script-based paths return complete data. Do NOT use broad keyword grep (`error|fail|...`): it matches cleanup steps, variable names, and incidental output\n- **OIDC-based workflows fail when the PR modifies the workflow file.** A forge limitation (provider reference \"Checks and CI logs\"); classify as informational when it applies\n- **Resolving the PR without a number is fragile.** Resolve `\u003Cpr_number>` once at phase entry and pass it to every later forge call (\"PR identity resolution\" above)\n- **Squash merge needs `git branch -D`, not `-d`.** After a squash merge the local branch commit is not in the default branch's history (different SHA), so `-d` says \"not fully merged.\" `-D` is safe because you already confirmed the merge\n- **`git add -A` and `git add .` are banned.** They risk committing secrets, build artifacts, or unrelated changes. Always `git add \u003Cspecific-files>`\n- **Bot comments need reactions AND replies.** React (👍\u002F👎\u002F👀) on every substantive comment AND reply with a per-finding classification table with evidence, for every reviewer. Verify BOTH the reaction and the reply landed on the forge by reading them back. Resolve the thread only when D7.5's conditions are met for EVERY finding in it, and only IF bot-authored; never human-authored, never your own. D7.5 in [reference\u002Fmonitor.md](reference\u002Fmonitor.md)'s per-iteration checklist is the single statement of which dispositions qualify and who may act on one, do not restate it here\n- **Monitor is async, not serial.** Process comments as they arrive while CI is still running; bots post at different times\n- **Check mergeable BEFORE polling CI.** While the PR's mergeability reports a conflict, the forge runs no workflows. Integrate the default branch first, then poll\n- **A green review lane can mean nothing was reviewed.** These lanes report on their session, not their output: a session that ends without error goes green even when it posted no review and no finding. No guard beside the lane judges what it posted; B2 ([reference\u002Fmonitor.md](reference\u002Fmonitor.md)'s per-iteration checklist) is the only productivity check, and billing is no signal, since the session is billed either way. Judge a review lane by what it produced (B2), never by its check row, and run the review locally when it produced nothing\n- **Never merge with unclassified FAILURE check runs.** Every FAILURE must be investigated, classified (real failure vs informational), and documented before merge is even suggested. See [readiness.md](reference\u002Freadiness.md) for the full 6-gate checklist\n- **\"No comments\" does NOT mean \"ready.\"** Comment-only actors post at unpredictable times; a 2-minute cooldown after the last check-run completion or comment arrival prevents the race. See readiness.md Gate 5\n- **Security scans are always blocking.** Any check run or bot comment reporting security findings (secrets, vulnerabilities) triggers mandatory triage; even a false positive must be explicitly classified and documented before merge\n- **Discover actors, don't hardcode them.** Security tools and AI reviewers change. Monitor discovers actors from the PR's checks and comments, classifies them by category (CI, security, review), and evaluates accordingly\n- **Uncommitted changes are silently lost on branch deletion.** `git reflog` recovers only commits. Before staging (Phase 2.3.1) and before the post-merge worktree transition (Phase 4.3), check `git status --porcelain` for unrelated uncommitted changes and stash them (`git stash push -m \"desc\" -- \u003Cfiles>`); stashes survive branch deletion. Never silently ignore uncommitted changes\n- **Cloud sessions poll.** Autonomous cloud sessions (`CLAUDE_CODE_REMOTE=true`) poll the PR's checks and comment surfaces on a fixed 60-90s cadence, and subscribe to PR activity where offered so reviews and failed checks wake them sooner. The poll stays, since pushes and green checks produce no event (§3.0.0 of monitor.md holds the decision, its pointer and its recheck trigger)\n- **Monitor checks the push channel first, then falls back.** Three tiers on local CLI sessions: (1) **push channel** (when your environment ships a forge-events MCP channel; ~0 idle requests), (2) **Monitor tool** (a `Monitor` watch armed with an explicit `timeout_ms` and re-armed at each deadline, per monitor.md §3.0.1; 60s poll fires on real CI\u002Fcomment events; cancel via `TaskStop`), (3) fixed-interval cron polling (deprecated, wasteful). Check for a channel before the Monitor tool: polling spends ~1 request per 30s interval vs ~0 idle with push delivery\n",{"metadata":24,"description":7,"argument-hint":27,"user-invocable":28,"disable-model-invocation":20},{"summary":25,"workflow-stage":26},"Full PR lifecycle. Prep, create, monitor CI, address reviews, merge","pr","[prep|create|ready|monitor|comments|ci|merge|reassess|evidence|status|full|…] [--unattended] [args]",true,"main","4f2dcaaff35e8f9fccc6a9f4d82a5830ab17a5c5","plugins\u002Fsource-control\u002Fskills\u002Fpull-request",7,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins","https:\u002F\u002Fgithub.com\u002Fmelodic-software","https:\u002F\u002Fmelodicsoftware.com","Independent software studio building developer tools, automation, and application services, with roots in music.","2026-11-15 05:15:06",{"source":8,"skillName":5,"commands":39,"prompt":49,"zipAvailable":28},[40,43,46],{"tool":41,"command":42},"npx","npx skills add melodic-software\u002Fclaude-code-plugins --skill pull-request",{"tool":44,"command":45},"bunx","bunx skills add melodic-software\u002Fclaude-code-plugins --skill pull-request",{"tool":47,"command":48},"pnpm","pnpm dlx skills add melodic-software\u002Fclaude-code-plugins --skill pull-request","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002FSKILL.md",[51,57,61,65,69,73,77,80,84,88,92,96,100,104,108,112,116,120,124,128,132,136,140,144,148,152,156,160,164,168,172,176,180,184,190,194,198,202,206,210,214,218,222,226,230,234,238,242,246,250,254,258,262,266,270,274,278,282,286,290,294,298,302,306,310,314,318,322,326,330,334,338,342,346,350,354,358,362,366,370,374,378,382,386,390,394,398,402,406,410,414,418,422],{"path":52,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":55,"sourceHtmlUrl":56},".claude\u002Fsource-control.md","REFERENCE","参考文档",562,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002F.claude\u002Fsource-control.md",{"path":58,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":59,"sourceHtmlUrl":60},"evals\u002Fdocs\u002Fconventions\u002Fsource-control.yaml",395,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fevals\u002Fdocs\u002Fconventions\u002Fsource-control.yaml",{"path":62,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":63,"sourceHtmlUrl":64},"evals\u002Fevals.json",89684,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fevals\u002Fevals.json",{"path":66,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":67,"sourceHtmlUrl":68},"evals\u002Ffixtures\u002Fci-status-security-fail.md",715,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fevals\u002Ffixtures\u002Fci-status-security-fail.md",{"path":70,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":71,"sourceHtmlUrl":72},"evals\u002Ffixtures\u002Fsource-control-open-draft.yaml",122,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fevals\u002Ffixtures\u002Fsource-control-open-draft.yaml",{"path":74,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":75,"sourceHtmlUrl":76},"evals\u002Ffixtures\u002Fsource-control-open-invalid.yaml",131,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fevals\u002Ffixtures\u002Fsource-control-open-invalid.yaml",{"path":78,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":71,"sourceHtmlUrl":79},"evals\u002Ffixtures\u002Fsource-control-open-ready.yaml","https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fevals\u002Ffixtures\u002Fsource-control-open-ready.yaml",{"path":81,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":82,"sourceHtmlUrl":83},"evals\u002Ffixtures\u002Fsource-control-pre-ready-security-off.yaml",211,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fevals\u002Ffixtures\u002Fsource-control-pre-ready-security-off.yaml",{"path":85,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":86,"sourceHtmlUrl":87},"evals\u002Ffixtures\u002Fsource-control-pre-ready-unknown-step.yaml",198,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fevals\u002Ffixtures\u002Fsource-control-pre-ready-unknown-step.yaml",{"path":89,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":90,"sourceHtmlUrl":91},"evals\u002Ffixtures\u002Fsource-control-prep-audits-replace-markdown.yaml",270,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fevals\u002Ffixtures\u002Fsource-control-prep-audits-replace-markdown.yaml",{"path":93,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":94,"sourceHtmlUrl":95},"evals\u002Ffixtures\u002Fsource-control-prep-audits-unknown-class.yaml",206,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fevals\u002Ffixtures\u002Fsource-control-prep-audits-unknown-class.yaml",{"path":97,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":98,"sourceHtmlUrl":99},"evals\u002Ffixtures\u002Fsource-control-required-none.yaml",195,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fevals\u002Ffixtures\u002Fsource-control-required-none.yaml",{"path":101,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":102,"sourceHtmlUrl":103},"evals\u002Ffixtures\u002Fsource-control-required-related.yaml",228,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fevals\u002Ffixtures\u002Fsource-control-required-related.yaml",{"path":105,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":106,"sourceHtmlUrl":107},"evals\u002Ffixtures\u002Fsource-control-required-summary-test-plan.yaml",221,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fevals\u002Ffixtures\u002Fsource-control-required-summary-test-plan.yaml",{"path":109,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":110,"sourceHtmlUrl":111},"reference\u002Fci-ci.md",10599,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Fci-ci.md",{"path":113,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":114,"sourceHtmlUrl":115},"reference\u002Fcomments-ci.md",11962,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Fcomments-ci.md",{"path":117,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":118,"sourceHtmlUrl":119},"reference\u002Fconflicts-ci.md",3096,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Fconflicts-ci.md",{"path":121,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":122,"sourceHtmlUrl":123},"reference\u002Fcreate.md",70529,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Fcreate.md",{"path":125,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":126,"sourceHtmlUrl":127},"reference\u002Fdependency-ci.md",7458,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Fdependency-ci.md",{"path":129,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":130,"sourceHtmlUrl":131},"reference\u002Fevidence.md",12014,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Fevidence.md",{"path":133,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":134,"sourceHtmlUrl":135},"reference\u002Ffull-lifecycle.md",3239,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Ffull-lifecycle.md",{"path":137,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":138,"sourceHtmlUrl":139},"reference\u002Fmedia.md",11358,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Fmedia.md",{"path":141,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":142,"sourceHtmlUrl":143},"reference\u002Fmerge.md",19709,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Fmerge.md",{"path":145,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":146,"sourceHtmlUrl":147},"reference\u002Fmonitor.md",75383,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Fmonitor.md",{"path":149,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":150,"sourceHtmlUrl":151},"reference\u002Fnative-surfaces.md",4200,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Fnative-surfaces.md",{"path":153,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":154,"sourceHtmlUrl":155},"reference\u002Fprep.md",13840,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Fprep.md",{"path":157,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":158,"sourceHtmlUrl":159},"reference\u002Fread.md",4026,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Fread.md",{"path":161,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":162,"sourceHtmlUrl":163},"reference\u002Freadiness.md",20628,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Freadiness.md",{"path":165,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":166,"sourceHtmlUrl":167},"reference\u002Fready-for-review.md",46817,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Fready-for-review.md",{"path":169,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":170,"sourceHtmlUrl":171},"reference\u002Freassess.md",7207,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Freassess.md",{"path":173,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":174,"sourceHtmlUrl":175},"reference\u002Freview-venue.md",7864,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Freview-venue.md",{"path":177,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":178,"sourceHtmlUrl":179},"reference\u002Freviewer-shapes.md",9412,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Freviewer-shapes.md",{"path":181,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":182,"sourceHtmlUrl":183},"reference\u002Fwatch-handoff.md",9256,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Freference\u002Fwatch-handoff.md",{"path":185,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":188,"sourceHtmlUrl":189},"scripts\u002Fcheck-body-sections.sh","SCRIPT","脚本",935,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fcheck-body-sections.sh",{"path":191,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":192,"sourceHtmlUrl":193},"scripts\u002Fcheck-body-sections.test.sh",4509,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fcheck-body-sections.test.sh",{"path":195,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":196,"sourceHtmlUrl":197},"scripts\u002Fcheck-gate-lines.sh",4008,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fcheck-gate-lines.sh",{"path":199,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":200,"sourceHtmlUrl":201},"scripts\u002Fcheck-gate-lines.test.sh",5021,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fcheck-gate-lines.test.sh",{"path":203,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":204,"sourceHtmlUrl":205},"scripts\u002Fcheck-ready-holds.sh",4339,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fcheck-ready-holds.sh",{"path":207,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":208,"sourceHtmlUrl":209},"scripts\u002Fcheck-ready-holds.test.sh",8763,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fcheck-ready-holds.test.sh",{"path":211,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":212,"sourceHtmlUrl":213},"scripts\u002Fconfirm-verdict.sh",6386,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fconfirm-verdict.sh",{"path":215,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":216,"sourceHtmlUrl":217},"scripts\u002Ffeedback-thread.sh",15946,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Ffeedback-thread.sh",{"path":219,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":220,"sourceHtmlUrl":221},"scripts\u002Ffeedback-thread.test.sh",30200,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Ffeedback-thread.test.sh",{"path":223,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":224,"sourceHtmlUrl":225},"scripts\u002Ffetch-annotations.sh",5835,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Ffetch-annotations.sh",{"path":227,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":228,"sourceHtmlUrl":229},"scripts\u002Ffetch-annotations.test.sh",15388,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Ffetch-annotations.test.sh",{"path":231,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":232,"sourceHtmlUrl":233},"scripts\u002Ffetch-failed-logs.sh",14433,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Ffetch-failed-logs.sh",{"path":235,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":236,"sourceHtmlUrl":237},"scripts\u002Ffetch-failed-logs.test.sh",16836,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Ffetch-failed-logs.test.sh",{"path":239,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":240,"sourceHtmlUrl":241},"scripts\u002Ffixtures\u002Fbody-with-reviewer-media.md",702,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Ffixtures\u002Fbody-with-reviewer-media.md",{"path":243,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":244,"sourceHtmlUrl":245},"scripts\u002Fmedia.py",37340,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fmedia.py",{"path":247,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":248,"sourceHtmlUrl":249},"scripts\u002Fmedia.test.sh",37263,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fmedia.test.sh",{"path":251,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":252,"sourceHtmlUrl":253},"scripts\u002Fmonitor-poll.test.sh",10681,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fmonitor-poll.test.sh",{"path":255,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":256,"sourceHtmlUrl":257},"scripts\u002Fparse-branch-issue.sh",13986,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fparse-branch-issue.sh",{"path":259,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":260,"sourceHtmlUrl":261},"scripts\u002Fparse-branch-issue.test.sh",26360,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fparse-branch-issue.test.sh",{"path":263,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":264,"sourceHtmlUrl":265},"scripts\u002Fparse-ready-skip.sh",2481,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fparse-ready-skip.sh",{"path":267,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":268,"sourceHtmlUrl":269},"scripts\u002Fparse-ready-skip.test.sh",4279,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fparse-ready-skip.test.sh",{"path":271,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":272,"sourceHtmlUrl":273},"scripts\u002Fpr-worktree.sh",3233,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fpr-worktree.sh",{"path":275,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":276,"sourceHtmlUrl":277},"scripts\u002Fpr-worktree.test.sh",7719,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fpr-worktree.test.sh",{"path":279,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":280,"sourceHtmlUrl":281},"scripts\u002Fpre-ready-steps.sh",354,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fpre-ready-steps.sh",{"path":283,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":284,"sourceHtmlUrl":285},"scripts\u002Fpre-ready-steps.test.sh",1271,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fpre-ready-steps.test.sh",{"path":287,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":288,"sourceHtmlUrl":289},"scripts\u002Fprep_warnings.py",6325,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fprep_warnings.py",{"path":291,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":292,"sourceHtmlUrl":293},"scripts\u002Fprep-warnings.sh",2525,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fprep-warnings.sh",{"path":295,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":296,"sourceHtmlUrl":297},"scripts\u002Fprep-warnings.test.sh",6852,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fprep-warnings.test.sh",{"path":299,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":300,"sourceHtmlUrl":301},"scripts\u002Fpush-branch.sh",24663,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fpush-branch.sh",{"path":303,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":304,"sourceHtmlUrl":305},"scripts\u002Fpush-branch.test.sh",60665,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fpush-branch.test.sh",{"path":307,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":308,"sourceHtmlUrl":309},"scripts\u002Fread-pr.sh",9376,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fread-pr.sh",{"path":311,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":312,"sourceHtmlUrl":313},"scripts\u002Fread-pr.test.sh",13850,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fread-pr.test.sh",{"path":315,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":316,"sourceHtmlUrl":317},"scripts\u002Fread-pre-ready.sh",6173,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fread-pre-ready.sh",{"path":319,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":320,"sourceHtmlUrl":321},"scripts\u002Fread-pre-ready.test.sh",11305,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fread-pre-ready.test.sh",{"path":323,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":324,"sourceHtmlUrl":325},"scripts\u002Fread-prep-audits.sh",5898,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fread-prep-audits.sh",{"path":327,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":328,"sourceHtmlUrl":329},"scripts\u002Fread-prep-audits.test.sh",9225,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fread-prep-audits.test.sh",{"path":331,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":332,"sourceHtmlUrl":333},"scripts\u002Fread-review-comments.sh",9055,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fread-review-comments.sh",{"path":335,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":336,"sourceHtmlUrl":337},"scripts\u002Fread-review-comments.test.sh",16916,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fread-review-comments.test.sh",{"path":339,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":340,"sourceHtmlUrl":341},"scripts\u002Fread-review-venue.sh",4346,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fread-review-venue.sh",{"path":343,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":344,"sourceHtmlUrl":345},"scripts\u002Fread-review-venue.test.sh",5196,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fread-review-venue.test.sh",{"path":347,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":348,"sourceHtmlUrl":349},"scripts\u002Fready-flip.test.sh",14940,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fready-flip.test.sh",{"path":351,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":352,"sourceHtmlUrl":353},"scripts\u002Fready-skip-comment.sh",2548,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fready-skip-comment.sh",{"path":355,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":356,"sourceHtmlUrl":357},"scripts\u002Fready-skip-comment.test.sh",3230,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fready-skip-comment.test.sh",{"path":359,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":360,"sourceHtmlUrl":361},"scripts\u002Fready-verify-local.test.sh",1693,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fready-verify-local.test.sh",{"path":363,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":364,"sourceHtmlUrl":365},"scripts\u002Freassess-comment.sh",2059,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Freassess-comment.sh",{"path":367,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":368,"sourceHtmlUrl":369},"scripts\u002Freassess-comment.test.sh",2637,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Freassess-comment.test.sh",{"path":371,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":372,"sourceHtmlUrl":373},"scripts\u002Freassess-limits.sh",7927,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Freassess-limits.sh",{"path":375,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":376,"sourceHtmlUrl":377},"scripts\u002Freassess-limits.test.sh",11149,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Freassess-limits.test.sh",{"path":379,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":380,"sourceHtmlUrl":381},"scripts\u002Fresolve-remote.sh",4247,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fresolve-remote.sh",{"path":383,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":384,"sourceHtmlUrl":385},"scripts\u002Fresolve-remote.test.sh",8096,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fresolve-remote.test.sh",{"path":387,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":388,"sourceHtmlUrl":389},"scripts\u002Freview-venue.sh",11989,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Freview-venue.sh",{"path":391,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":392,"sourceHtmlUrl":393},"scripts\u002Freview-venue.test.sh",24193,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Freview-venue.test.sh",{"path":395,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":396,"sourceHtmlUrl":397},"scripts\u002Fsiblings-line.sh",2751,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fsiblings-line.sh",{"path":399,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":400,"sourceHtmlUrl":401},"scripts\u002Fsiblings-line.test.sh",3215,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fsiblings-line.test.sh",{"path":403,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":404,"sourceHtmlUrl":405},"scripts\u002Ftarget-repo.sh",3884,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Ftarget-repo.sh",{"path":407,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":408,"sourceHtmlUrl":409},"scripts\u002Ftarget-repo.test.sh",5938,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Ftarget-repo.test.sh",{"path":411,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":412,"sourceHtmlUrl":413},"scripts\u002Fverify-gate.sh",3532,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fverify-gate.sh",{"path":415,"resourceType":186,"resourceTypeDesc":187,"sizeBytes":416,"sourceHtmlUrl":417},"scripts\u002Fverify-gate.test.sh",4614,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Fscripts\u002Fverify-gate.test.sh",{"path":419,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":420,"sourceHtmlUrl":421},"templates\u002Fchecklist.md",2114,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Ftemplates\u002Fchecklist.md",{"path":423,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":424,"sourceHtmlUrl":425},"templates\u002Fevidence-comment.md",329,"https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsource-control\u002Fskills\u002Fpull-request\u002Ftemplates\u002Fevidence-comment.md",[427,434,441,448,453,460],{"skillNo":428,"name":429,"nameZh":6,"description":430,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":431,"categories":432,"keywords":433,"publishedAt":19},"01M7DYYDG31B9YX7N9MB9QW2GW","track","Backlog reads through the bound tracker: search open and closed items, view one item with its comments and state history. Use when another skill searches or reads tracker items.","https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fsession-flow\u002Fevals\u002Frecall-still-reports-reopened-bug\u002Fwork-items-stub\u002Fskills\u002Ftrack\u002FSKILL.md",[],[],{"skillNo":435,"name":436,"nameZh":6,"description":437,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":438,"categories":439,"keywords":440,"publishedAt":19},"01M7DYXVBAVG3NV30PA7GKW9Q6","reviewer-media","Reviewer media for one pull request: decides whether it needs a narrated walkthrough, builds the storyboard from \u002Freview:explain-change's digest, has \u002Fexplainer-video:produce render it and hands the manifest to \u002Fsource-control:pull-request media. Use when asked for 'reviewer media', 'a walkthrough video for this PR', 'a produced video of this change', or when pull-request ready or monitor hands a PR over.","https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Freview\u002Fskills\u002Freviewer-media\u002FSKILL.md",[],[],{"skillNo":442,"name":443,"nameZh":6,"description":444,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":445,"categories":446,"keywords":447,"publishedAt":19},"01M7DYRZ5YD44AMYH26YR25GZ8","route","Model and effort per agent role (orchestrator, worker, verifier, retrieval) for a multi-agent run, resolved through the bundled defaults and the user, team and personal override config layers into fenced JSON with single-agent and fan-out variants and the layer behind each value. Use when asked 'which model for the subagents', 'route the roles' or 'model routing', or before a skill passes a role's model on a spawn.","https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fdiscovery\u002Fevals\u002Fex-model-pin-without-override\u002Fmulti-agent-stub\u002Fskills\u002Froute\u002FSKILL.md",[],[],{"skillNo":449,"name":443,"nameZh":6,"description":444,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":450,"categories":451,"keywords":452,"publishedAt":19},"01M7DYRTVT4FWTGDJY0HCCFJMZ","https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fdiscovery\u002Fevals\u002Fex-model-override-from-route\u002Fmulti-agent-stub\u002Fskills\u002Froute\u002FSKILL.md",[],[],{"skillNo":454,"name":455,"nameZh":6,"description":456,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":457,"categories":458,"keywords":459,"publishedAt":19},"01M7DYNZC8PNW3T27V2ADKECKJ","principles","Module-design vocabulary for architecture findings, and the checks built on it. Use when: 'what is a deep module', 'what is a seam', 'deep vs shallow module', 'what is the deletion test', 'leverage and locality', 'module design vocabulary', 'which word do we use for X in architecture findings'. Scanning a codebase for friction is \u002Farchitecture:improve.","https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Farchitecture\u002Fskills\u002Fprinciples\u002FSKILL.md",[],[],{"skillNo":461,"name":462,"nameZh":6,"description":463,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":464,"categories":465,"keywords":466,"publishedAt":19},"01M7DYNMCZNEWBFHV9KTSS6H60","write","Structured five-field bug report (title, steps to reproduce, expected vs actual, justified severity, suggested fix location) from an informal description, read-only. Use when the user names a defect they observed, or asks for one written up ('report a bug', 'file a bug', 'write this up as a bug').","https:\u002F\u002Fgithub.com\u002Fmelodic-software\u002Fclaude-code-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Farchitecture\u002Fevals\u002Fdeepening-defect-goes-to-bugs-write\u002Fbugs-stub\u002Fskills\u002Fwrite\u002FSKILL.md",[],[]]