Name: apm-review-panel
Rating: 67.2 (1 reviews)
Author: microsoft

apm-review-panel

Use this skill to run a multi-persona expert advisory review on a labelled pull request in microsoft/apm. The panel fans out to five mandatory specialists plus a test-coverage specialist (active on every PR that touches src/) plus three conditional specialists (auth, doc-writer, performance-expert), all running in their own agent threads, and a CEO synthesizer. The orchestrator is the sole writer to the PR: ONE recommendation comment, no verdict labels, no merge gating. The panel is advisory -- it surfaces findings, prioritizes follow-ups, and renders a ship-recommendation that the maintainer and author weigh. Activate when a non-trivial PR needs a cross-cutting recommendation (architecture, CLI logging, DevX UX, supply-chain security, growth/positioning, optionally auth, docs, perf, and test coverage, with CEO arbitration).

Quality

83%

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

73%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.

The body is highly actionable with a clear, well-validated workflow and well-signaled references, but it is markedly verbose: key invariants are restated three times and large lookup tables are inlined rather than externalized.

Suggestions

Consolidate the 'non-empty turn exit' and 'synchronous fan-out' rules into a single architecture invariant and reference it from the checklist and gotchas instead of restating them near-verbatim three times.

Move the long conditional-activation file lists (auth, doc-writer, performance-expert) into a reference file and link to it, keeping only the activation criteria and fallback self-checks inline.

Tighten the Gotchas section by removing entries that re-explain invariants already covered above, retaining only the genuinely novel traps (e.g. the mermaid :::cssClass shorthand).

Dimension	Reasoning	Score
Conciseness	At ~511 lines the body reiterates the same load-bearing rules (non-empty turn exit, synchronous fan-out, single-writer interlock) across the architecture invariants, execution checklist, and gotchas, often near-verbatim -- noticeably verbose and padded.	2 / 5
Actionability	Provides concrete, executable guidance: the exact `gh pr view` command, a literal JSON placeholder for schema failures, the exact label list `[panel-review, panel-approved, panel-rejected]`, schema citations, and step-by-step task-prompt clauses.	5 / 5
Workflow Clarity	A clear 9-step sequence with explicit validation checkpoints (S4 schema gate with bounded re-spawns, CEO return validation, step-9 non-empty-exit contract) and feedback loops for the single-writer batch/label-sweep operations.	5 / 5
Progressive Disclosure	Well-structured with one-level-deep references to real assets (panelist/CEO schemas, recommendation template, persona .agent.md files), though large conditional-activation file lists and the routing matrix are inlined where they could be split out.	4 / 5
	Total	16 / 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.

The description is specific, complete, and distinctive, with an explicit activation clause and concrete actions. It is slightly weighted toward internal jargon over everyday trigger terms but is otherwise strong.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions -- fans out to named specialists, schema-validates returns, runs a CEO synthesizer, renders one recommendation comment, and sweeps labels -- covering the workflow comprehensively.	5 / 5
Completeness	Explicitly answers both 'what' (multi-persona advisory review with fan-out + CEO synthesizer, single-writer to the PR) and 'when' via a concrete 'Activate when a non-trivial PR needs a cross-cutting recommendation' clause.	5 / 5
Trigger Term Quality	Includes natural trigger phrases like 'advisory review', 'labelled pull request', and 'non-trivial PR' alongside domain terms, but leans on jargon ('DevX UX', 'growth/positioning') and omits a few common synonyms.	4 / 5
Distinctiveness Conflict Risk	Targets a narrow, well-defined niche (labelled PR in microsoft/apm, multi-persona advisory panel with CEO arbitration) with distinct triggers and minimal overlap with other skills.	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
skill_md_line_count	SKILL.md is long (512 lines); consider splitting into references/ and linking	Warning
relative_links	Relative link issues: 10 suspicious	Warning
referenced_paths_exist	Referenced path issues: 2 missing	Warning

	Total	13 / 16 Passed

Repository: microsoft/apm
Path: .agents/skills/apm-review-panel/SKILL.md
Commit: 119aadd

Reviewed: 5 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.