CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

issue-triage

3-phase issue backlog management with audit, deep analysis, and validated triage actions. Use when triaging GitHub issues, sorting bug reports, cleaning up stale tickets, or detecting duplicate issues. Args: 'all' to analyze all, issue numbers to focus (e.g. '42 57'), 'en'/'fr' for language, no arg = audit only.

Quality

86%

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 1-5

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

An exceptionally actionable and well-sequenced workflow skill: executable commands, explicit validation gates before any destructive action, protection rules, fallbacks, and thorough edge-case handling. Its weaknesses are structural: it is monolithic with no progressive disclosure, references a templates/issue-comment.md file that is missing from the bundle, and contains minor duplicated notes that could be trimmed.

Suggestions

Create the referenced templates/issue-comment.md file (or remove the Phase 3 reference to it) — as written, the skill points to a bundle file that does not exist.

Move the bulky, rarely-needed-per-run material — the output triage-table templates, the full Jaccard algorithm spec, and the /pr-triage comparison table — into a references/ file linked one level deep, keeping SKILL.md as a leaner overview of the 3 phases.

De-duplicate the repeated guidance (the 'comments field is count-only' note appears in both Phase 1 and Notes; the English-only comment rule appears in both Language and Phase 3) to reclaim tokens.

DimensionReasoningScore

Conciseness

The body is dense and operational — exact gh commands, decision tables, and a self-contained Jaccard algorithm with no explanations of concepts Claude already knows. Not 5 due to minor duplication and padding: the Notes repeat the Phase 1 "comments field returns count, not content" note verbatim, Phase 3 re-states the English-language rule from the Language section, and the closing /pr-triage comparison table largely restates the decision rule; not 3 because these are trims, not conceptual over-explanation.

4 / 5

Actionability

Quotes: "gh issue list --state open --limit 100 --json number,title,author,...", "Patterns: `fixes #N`, `closes #N`, `resolves #N`...", the 4-step Jaccard algorithm with stop-word list and thresholds, exact output table templates, AskUserQuestion option blocks, and copy-paste execution heredocs ("gh issue comment {num} --body-file - <<'TRIAGE_EOF'"). Fully executable guidance with edge cases and fallbacks covering the common cases.

5 / 5

Workflow Clarity

The 3 phases are clearly sequenced with preconditions ("git rev-parse --is-inside-work-tree; gh auth status"), and the destructive/batch operations have explicit validation: "Display ALL drafted actions" before an AskUserQuestion gate, "Never execute any action without explicit user validation in chat", plus protection rules, fallbacks (collaborators 403 → merged-PR authors; parallel → sequential), and an edge-case table with recovery behavior. The batch-operation cap at 3 does not apply because validation checkpoints are present.

5 / 5

Progressive Disclosure

The skill is a single ~450-line monolithic SKILL.md with no bundle files, yet Phase 3 references "the template `templates/issue-comment.md`" — a file that does not exist in the bundle (dangling reference). Internal headers and tables give it clear structure, so anchor 2's "minimal structure" does not fit, but content that could live one level deep (output table templates, the Jaccard algorithm spec, the /pr-triage comparison) is fully inlined and the one reference made is broken, so it does not reach anchor 4's "most content appropriately placed; references mostly clear".

3 / 5

Total

17

/

20

Passed

Description

83%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: it answers both what (3-phase audit/analysis/validated triage actions) and when (explicit 'Use when' clause with natural trigger phrases and synonym coverage), and documents argument handling. Minor improvements possible by enumerating the concrete triage actions (comment/label/close) and adding a few more natural trigger variations.

DimensionReasoningScore

Specificity

Quotes: "audit, deep analysis, and validated triage actions" plus "sorting bug reports, cleaning up stale tickets, or detecting duplicate issues" — several named actions with only minor coverage gaps (comment/label/close actions are not enumerated in the description itself). Not 5 because "audit" and "deep analysis" are somewhat abstract compared to the fully concrete action lists in the top anchor; not 3 because it goes well beyond 1-2 actions.

4 / 5

Completeness

Quotes: "3-phase issue backlog management with audit, deep analysis, and validated triage actions" (clear what) and "Use when triaging GitHub issues, sorting bug reports, cleaning up stale tickets, or detecting duplicate issues" (explicit when with concrete trigger phrases). Matches the top anchor exactly; the when-clause is explicit, not merely implied, so 4 does not fit.

5 / 5

Trigger Term Quality

Quotes: "triaging GitHub issues", "sorting bug reports", "cleaning up stale tickets", "detecting duplicate issues", "backlog" — natural phrasings with synonym coverage (issues/tickets). Not 5 because common variations like "labeling issues", "closing issues", or "issue cleanup" phrasings are absent; not 3 because coverage is broad and natural, not partial.

4 / 5

Distinctiveness Conflict Risk

Quotes: "GitHub issues", "issue backlog" — a clear niche with distinct triggers. Not 5 because "sorting bug reports" is generic and a closely related sibling skill (/pr-triage) creates minor overlap risk; not 3 because the scope is firmly pinned to issue triage, not merely "somewhat specific".

4 / 5

Total

17

/

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

FlorianBruniaux/claude-code-ultimate-guide

Path

examples/skills/issue-triage/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.