Name: autopilot-pr-merge-worker
Rating: 71.2 (1 reviews)
Author: microsoft

autopilot-pr-merge-worker

Drive ONE already selected open pull request in microsoft/apm to mergeable. Classifies copilot-pull-request-reviewer[bot] inline review, runs autopilot-pr-review-worker, folds (by default) every recommendation inside the PR's stated scope, pushes to the head branch or a superseding PR that preserves authorship via commit trailers, watches CI to green, and iterates under fixed caps until ready-to-merge, advisory-with-deferred, superseded, or blocked. Never spawned by autopilot-pr-review-scheduler. That scheduler is advisory only. Invoke this skill by name when the caller asked to drive a PR to merge.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

85%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, actionable skill body with a clear phased workflow, embedded validation gates, and clean progressive disclosure to verified bundle files. Slight gains available from de-duplicating repeated boundary statements and inlining a few more concrete command examples.

Suggestions

De-duplicate the repeated boundary claims (fold-by-default, scheduler-never-spawns, owner-gate fail-closed) into a single canonical section to trim tokens.

Inline one or two more concrete commands (e.g. the owner_touch_gate.py invocation and completion-schema validation command) instead of only pointing at the bundled assets.

Briefly gloss the load-bearing acronyms (A9 SUPERVISED EXECUTION, HAND-ROLLED HALLUCINATION, PHANTOM DEPENDENCY) on first use so the terms are not opaque to a fresh reader.

Dimension	Reasoning	Score
Conciseness	Mostly lean and assumes competence (unexplained A9/SUPERVISED EXECUTION tags, terse phase list), but several boundary statements are repeated across sections (fold-by-default, scheduler-never-spawns, owner-gate-as-FAIL-CLOSED) and could be tightened; efficient with minor over-explanation.	4 / 5
Actionability	Provides executable artifacts — the activation card template, a copy-paste `test -f ...` probe, and the concrete lint pair (`uv run --extra dev ruff check src/ tests/`, `ruff format --check src/ tests/`) — but the bulk of executable detail is delegated to bundled assets rather than inlined, leaving minor gaps.	4 / 5
Workflow Clarity	Phases X.0 through X.8 are explicitly sequenced with validation checkpoints baked in (canonical-owner gate at X.2.5, lint contract until silent at X.4, CI-observed-green at X.6/X.8) plus feedback loops (CI recovery) and hard caps, satisfying the destructive/batch validation requirement.	5 / 5
Progressive Disclosure	Clear overview with one-level-deep, well-signaled references to real bundle files (assets/, references/, scripts/ all verified present on disk), each link annotated with its purpose and phase; content is appropriately split with easy navigation and no nested-reference chains.	5 / 5
	Total	18 / 20 Passed

Description

92%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, specific description that covers concrete actions, explicit when-guidance, and clear boundary from sibling skills in third-person voice. Keyword coverage is good but could add a few more natural synonyms ('merge', 'land a PR').

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Classifies copilot-pull-request-reviewer[bot] inline review', 'runs autopilot-pr-review-worker', 'folds (by default) every recommendation', 'pushes to the head branch or a superseding PR', 'watches CI to green', 'iterates under fixed caps' — giving comprehensive, specific coverage rather than vague language.	5 / 5
Completeness	Explicitly answers both what (the full drive-to-merge convergence loop) and when — 'Invoke this skill by name when the caller asked to drive a PR to merge' is a concrete, explicit trigger phrase.	5 / 5
Trigger Term Quality	Has natural phrases a caller would say ('drive a PR to merge', 'pull request', 'mergeable', 'CI to green') but is missing common synonyms/variations like 'merge a PR' or 'land a PR'; good but not exhaustive keyword coverage.	4 / 5
Distinctiveness Conflict Risk	Carves a clear niche — a per-PR merge worker for microsoft/apm — and explicitly disambiguates from siblings ('Never spawned by autopilot-pr-review-scheduler. That scheduler is advisory only.'), minimizing conflict risk.	5 / 5
	Total	19 / 20 Passed

Validation

81%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 2 suspicious	Warning
referenced_paths_exist	Referenced path issues: 1 missing	Warning

	Total	13 / 16 Passed

Repository: microsoft/apm
Path: .agents/skills/autopilot-pr-merge-worker/SKILL.md
Commit: b914b1b

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.