CtrlK
BlogDocsLog inGet started
Tessl Logo

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).

73

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

85%Weight 40%Scale 1-3

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

A highly actionable, well-sequenced orchestration skill with real commands, schema gates, and properly externalized bundle files. Its main weakness is verbosity: several load-bearing invariants are restated many times and the body could be tightened substantially without losing information.

Suggestions

Consolidate the synchronous-fan-out / never-spawn-and-forget rule into a single authoritative statement; it is currently repeated across the architecture invariants, the checklist intro, step 5, and the Gotchas section.

De-duplicate the advisory/no-verdict-label and the [panel-review, panel-approved, panel-rejected] sweep points, which each recur 3-4 times across invariants, the output contract, step 8, and Gotchas.

Consider moving the long per-persona conditional-activation file-path lists (auth, doc-writer, performance-expert) into a referenced ROUTING.md so the body stays a lean overview.

DimensionReasoningScore

Conciseness

The ~490-line body restates key invariants multiple times (synchronous fan-out appears ~4x across invariants, the checklist intro, step 5, and Gotchas; the label sweep and advisory/no-verdict points recur 3-4x each), so it could be tightened even though it avoids explaining concepts Claude already knows. It is above anchor 1 (operational, not padded with basic concepts) but below anchor 3 (clear redundancy means not every token earns its place).

2 / 3

Actionability

Provides fully executable guidance: real commands ('gh pr view <PR_NUMBER> --json author,reviewRequests', 'safe-outputs.remove-labels', 'safe-outputs.add-comment.max: 2'), a concrete re-spawn policy with a placeholder JSON, and real referenced schemas, matching the copy-paste-ready anchor.

3 / 3

Workflow Clarity

A clearly sequenced 9-step checklist with explicit validation checkpoints (step 4 schema gate, step 5 CEO-return validation) and feedback loops (validate then re-spawn, max two attempts), plus the non-empty-turn-exit hard contract, matching the anchor for clear sequence with explicit validation.

3 / 3

Progressive Disclosure

Detailed material is externalized to verified, one-level-deep, clearly-signaled bundle files (assets/panelist-return-schema.json, assets/ceo-return-schema.json, assets/recommendation-template.md) cited by relative path; the body holds orchestration logic rather than reference material that should be split out.

3 / 3

Total

11

/

12

Passed

Description

100%Weight 40%Scale 1-3

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 states concrete capabilities, includes natural trigger terms, and explicitly covers both what the skill does and when to use it. It is clearly distinguishable from other skills.

DimensionReasoningScore

Specificity

Lists multiple concrete actions ('fans out to five mandatory specialists', 'CEO synthesizer', 'ONE recommendation comment', 'no verdict labels', 'prioritizes follow-ups', 'renders a ship-recommendation'), matching the anchor for multiple specific concrete actions rather than just naming a domain.

3 / 3

Completeness

Explicitly answers both what (multi-persona advisory review with fan-out + synthesizer, single comment) and when via an explicit 'Activate when a non-trivial PR needs a cross-cutting recommendation...' clause, so it is not capped at 2 for a missing trigger.

3 / 3

Trigger Term Quality

Covers natural terms a maintainer would say ('pull request', 'PR', 'recommendation', 'architecture', 'supply-chain security', 'auth', 'docs', 'perf', 'test coverage', 'CLI logging', 'DevX UX'), giving good coverage of common variations rather than only technical jargon.

3 / 3

Distinctiveness Conflict Risk

A very specific niche (multi-persona advisory panel on microsoft/apm PRs with a named specialist roster) with distinct triggers, making it unlikely to fire for the wrong skill.

3 / 3

Total

12

/

12

Passed

Validation

81%

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

Validation13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

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
Reviewed

Table of Contents

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.