CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

autopilot-comment

Use this skill to post or patch ONE GitHub comment for a microsoft/apm autopilot run. It owns public vs debug body assembly and the AI disclaimer footer. Never labels, assigns, requests reviewers, or merges. Autopilot skills that write comments must load this skill and must not call gh comment APIs themselves.

72

Quality

91%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

The canonical home for this skill is autopilot-comment in microsoft/apm

SKILL.md
Quality
Evals
Security

Quality

Content

88%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 a lean, fail-closed rule spec: a gated activation card, numbered assembly steps with an exact footer, concrete per-environment write commands, and explicit hard nos. The only gaps are minor actionability details (the body-file path and PATCH body parameters) and the absence of any reference structure.

Suggestions

Complete the write commands: specify how the body file is created (e.g., write to a temp file with the ASCII-only assembled body) and show the full PATCH invocation including the request body parameter (e.g., gh api -X PATCH ... -f body=@file).

Clarify the Agentic Workflow path with the same concreteness as the actor-session path — e.g., the exact safe-outputs.add-comment payload shape and what to do when the platform rejects it, instead of only 'stop'.

DimensionReasoningScore

Conciseness

The body is pure imperative rules with zero concept explanation ('One comment per activation. No progress comments.', 'ASCII only.') and assumes Claude's competence. The repetition of the assign/label/merge prohibition between card rules and 'Hard nos' is deliberate fail-closed enforcement, not padding, so it fits the 5 anchor rather than 4's 'could be trimmed'.

5 / 5

Actionability

Gives concrete commands ('gh issue comment <target> --body-file ...', 'gh api -X PATCH /repos/microsoft/apm/issues/comments/<comment_id>') and an exact copy-paste footer, but the --body-file target is an ellipsis and the PATCH request body is unspecified, so it is not fully copy-paste ready — anchor 4 rather than 5.

4 / 5

Workflow Clarity

Clear sequence (emit activation card → assemble → write → exit receipt) with explicit validation checkpoints ('Missing field -> stop', 'Mismatch -> stop', footer dedup check). This is a single unambiguous write of one comment, not a batch or destructive operation, so the simple-skill exception applies and 5 fits.

5 / 5

Progressive Disclosure

Sections (Activation card, Assemble, Write, Hard nos) are well organized and everything is appropriately inline for a compact rule spec, but no bundle files exist and the ~100-line body exceeds the under-50-line simple-skill line, so it matches anchor 4's 'good structure, minor organization gaps' rather than 5's well-signaled references.

4 / 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.

The description is highly specific, third-person, and complete: it states what the skill does (post/patch exactly one comment, body assembly, disclaimer footer), when to load it, and what it never does. Trigger-term coverage is strong but misses a few natural synonyms like 'reply' or 'respond'.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — 'post or patch ONE GitHub comment', 'owns public vs debug body assembly and the AI disclaimer footer' — plus explicit exclusions ('Never labels, assigns, requests reviewers, or merges'), in third person. Coverage of its niche is comprehensive, so it fits the 5 anchor rather than 4's 'minor gaps'.

5 / 5

Completeness

'Use this skill to post or patch ONE GitHub comment for a microsoft/apm autopilot run' answers what, and 'Autopilot skills that write comments must load this skill' answers when with a concrete trigger phrase. Both are explicit, matching the 5 anchor rather than 4's 'when could be more explicit'.

5 / 5

Trigger Term Quality

Natural phrases like 'post or patch', 'GitHub comment', 'autopilot run', and 'gh comment APIs' give good keyword coverage, but common user variations such as 'reply', 'respond', or comment URL forms are missing. Fits anchor 4 ('a few natural terms missing') rather than 5's comprehensive synonym coverage.

4 / 5

Distinctiveness Conflict Risk

The sole-writer niche ('microsoft/apm autopilot run', 'must not call gh comment APIs themselves') is distinct and explicitly de-conflicted with caller skills, giving minimal overlap risk. Clearly the 5 anchor, not 4's 'minor overlap risk'.

5 / 5

Total

19

/

20

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.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

microsoft/apm

Path

packages/autopilot/autopilot-comment/SKILL.md

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.