Name: triage-issue
Rating: 72 (1 reviews)
Author: docker

triage-issue

Analyze a single GitHub issue for docker/docs — check whether the problem still exists, determine a verdict, and report findings. Use when asked to triage, assess, or review an issue, even if the user doesn't say "triage" explicitly: "triage issue 1234", "is issue 500 still valid", "should we close #200", "look at this issue", "what's going on with #200".

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

82%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 highly actionable, well-sequenced triage workflow with real repo-specific knowledge (timeline cross-references, /manuals prefix mapping, upstream ownership rules) and zero filler. The main improvements are mechanical: deduplicate the repeated label-removal command, remove the stray trailing period, and add a light post-action verification step.

Suggestions

Factor the repeated 'gh api repos/docker/docs/issues/$ARGUMENTS/labels/status%2Ftriage --method DELETE || true' command out of the three verdict blocks into a single shared step (e.g. 'after applying any verdict label, remove status/triage') to reduce duplication (conciseness).

Add a post-action verification step, such as re-fetching the issue's labels/state after acting to confirm the close or label change landed, giving the workflow an explicit feedback loop (workflow_clarity).

Delete the stray '.' line at the end of the file and consider moving the area-label catalogue to a short reference file, keeping the decision flow inline (conciseness, progressive_disclosure).

Dimension	Reasoning	Score
Conciseness	The body is lean with zero concept padding — every section is commands or decision rules Claude does not already know (e.g. the '/manuals prefix mapping', the vendored-content ownership rule). It falls short of 5 because the 'gh api ... labels/status%2Ftriage --method DELETE \|\| true' command is duplicated verbatim across three verdict blocks, and a stray '.' line sits at the end of the file — both easy to trim.	4 / 5
Actionability	Every step is copy-paste-ready: the fetch specifies exact '--json' fields, the timeline check includes a complete jq filter, and each verdict has an exact 'gh issue close' or 'gh api labels' command. This matches the fully-executable top anchor with the common cases covered.	5 / 5
Workflow Clarity	A clear seven-step sequence (fetch, understand, follow URLs, check repo, upstream, decide, report) with explicit decision criteria ('A merged PR is strong evidence the issue is fixed') and four labeled verdicts. It misses 5 because there is no post-action verification or error-recovery loop (e.g. confirming labels applied or re-checking state after closing), and the 'gh issue close' action is only implicitly gated by the investigation.	4 / 5
Progressive Disclosure	Well-organized numbered sections, no nested or buried references, and the only external pointer (the vendored content table in CLAUDE.md) is clearly signaled. Not 5: the ~28-item area-label catalogue and the label-management commands are inlined where a small reference file would keep the decision flow tighter, and the skill exceeds the under-50-line simple-skill case.	4 / 5
	Total	17 / 20 Passed

Description

95%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 description: concrete, third-person, explicitly stating both what the skill does and when to use it, with unusually good example trigger phrasings including 'should we close #200' and 'is issue 500 still valid'. The only minor gap is that the description omits the concrete verdict actions (close, label, escalate) that the body delivers.

Dimension	Reasoning	Score
Specificity	Names the domain ('a single GitHub issue for docker/docs') and three concrete actions ('check whether the problem still exists, determine a verdict, and report findings'), matching the several-specific-actions-with-minor-gaps anchor. It stops short of 5 because the outcome actions the skill actually performs (closing the issue, labeling, escalating upstream) are absent from the description.	4 / 5
Completeness	Both what ('Analyze a single GitHub issue... check whether the problem still exists, determine a verdict, and report findings') and when ('Use when asked to triage, assess, or review an issue...') are explicitly stated with concrete trigger phrases, matching the top anchor. Not 4 because the when-clause is fully explicit rather than merely present.	5 / 5
Trigger Term Quality	'Use when asked to triage, assess, or review an issue' is backed by five natural example phrasings ('triage issue 1234', 'is issue 500 still valid', 'should we close #200', 'look at this issue', "what's going on with #200"), covering synonyms and paraphrases users would actually say. Nothing natural is missing for this domain, so the comprehensive-coverage anchor fits.	5 / 5
Distinctiveness Conflict Risk	The description is scoped to 'a single GitHub issue for docker/docs', a clear niche with distinct triggers (issue numbers, close/still-valid phrasings) and minimal overlap with generic review or issue-management skills. The repo-specific scoping is what distinguishes it from the anchor-4 case of closely related skills overlapping.	5 / 5
	Total	19 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: docker/docs
Path: .agents/skills/triage-issue/SKILL.md
Commit: 4d3cbcd

Reviewed: 1 day 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.