CtrlK
BlogDocsLog inGet started
Tessl Logo

apm-spec-guardian

Use this skill to run a four-panel adversarial advisory review on any pull request that touches the OpenAPM specification artifact (docs/src/content/docs/specs/openapm-*.md), its inline / sidecar JSON Schemas (docs/src/content/docs/specs/schemas/*.schema.json), or the conformance fixture seed (tests/fixtures/spec-conformance/**). The panel fans out to four spec-ecosystem reviewers (swagger-openapi-editor, oci-distribution-editor, pkgmgr-registry-contract-editor, w3c-tag-architect), each running in its own agent thread, and a spec-editor synthesizer that produces a fold-now / defer-v0.1.1 / defer-v0.2 / reject list plus a ship decision keyed off a 1..10 shocked_meter scale. The orchestrator is the sole writer to the PR: ONE consolidated comment, no verdict labels, no merge gating. The panel is advisory -- it surfaces findings, prioritizes folds, and renders a ship recommendation that the maintainer weighs.

69

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

High

Do not use without reviewing

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 well-engineered orchestration skill body: actionable, with a clearly sequenced multi-wave workflow, explicit validation checkpoints, and clean one-level-deep externalization of schemas, template, and linter. The only real weakness is conciseness — the advisory-regime point is restated several times and the topology diagram is large.

Suggestions

Consolidate the 'advisory, not gate / no verdict labels' point so it is stated once in 'Architecture invariants' and referenced rather than re-explained in Wave 4, Wave 6, the output contract, and Gotchas.

Consider condensing the ASCII topology diagram or moving the per-wave detail into the wave sections to reduce redundant rendering of the same structure.

DimensionReasoningScore

Conciseness

The body is dense and operational and assumes Claude's competence (no basic-concept padding), but it is long and restates the 'advisory, not gate' invariant across the invariants, Wave 4, Wave 6, output contract, and gotchas, and the ASCII topology could be tightened.

2 / 3

Actionability

Gives concrete executable guidance: exact asset paths, exact thresholds ('shocked_meter_avg >= 7.0', blocker veto, 'max 2 attempts'), itemized per-task prompt contents, and explicit ordered Wave 0 decision rules — copy-paste-ready for an orchestration skill.

3 / 3

Workflow Clarity

Waves 0–6 are clearly sequenced with explicit validation checkpoints (schema-validate each panelist return, validate the synthesizer return, 11-check linter), feedback loops (re-spawn on malformed returns), and defined loop budgets — matching the clear-sequence-with-validation anchor.

3 / 3

Progressive Disclosure

The body is an overview pointing to four well-signaled, one-level-deep asset files (panelist-return-schema.json, synthesizer-return-schema.json, linter-checklist.md, comment-template.md), all verified present, with the heavy schemas/template/linter appropriately externalized rather than inlined.

3 / 3

Total

11

/

12

Passed

Description

85%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 specific, complete, and distinctive description that clearly states what the skill does and when it fires. Its main weakness is trigger-term phrasing: the activation condition is given as repo path globs and internal artifact names rather than natural-language terms a maintainer would voice.

Suggestions

Add a plain-language trigger clause naming what a maintainer would say (e.g. 'spec review', 'OpenAPM spec change', 'conformance fixture update') alongside the path globs.

Surface a couple of common trigger-term variations (e.g. 'spec PR', 'schema change', 'fixture seed') so the description matches natural requests, not just path patterns.

DimensionReasoningScore

Specificity

Lists multiple concrete actions: 'fans out to four spec-ecosystem reviewers', 'produces a fold-now / defer-v0.1.1 / defer-v0.2 / reject list plus a ship decision keyed off a 1..10 shocked_meter scale', and 'ONE consolidated comment, no verdict labels, no merge gating'.

3 / 3

Completeness

Explicitly answers both what (four-panel adversarial advisory review plus synthesizer) and when ('on any pull request that touches the OpenAPM specification artifact ... or the conformance fixture seed'), which is an equivalent explicit 'Use when' trigger.

3 / 3

Trigger Term Quality

The trigger is expressed as path globs and internal artifact names ('openapm-*.md', '*.schema.json', 'tests/fixtures/spec-conformance/**') rather than natural-language variations a user would say; relevant terms are present but lean technical, missing common phrasings.

2 / 3

Distinctiveness Conflict Risk

Occupies a narrow spec-only niche gated by explicit path triggers, and is explicitly distinguished from its sibling 'apm-review-panel' (different roster, shocked_meter vs stance enum), making wrong-skill conflicts unlikely.

3 / 3

Total

11

/

12

Passed

Validation

93%

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

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 5 suspicious

Warning

Total

15

/

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.