CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

issue-triage

Scan open issues and discussions. Dedupe, prioritize, and propose labels. Updates issue-triage-state.md. L1 propose-only — never auto-label or close.

Quality

80%

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

Fix and improve this skill with Tessl

tessl review fix ./starters/issue-triage/.codex/skills/issue-triage/SKILL.md

The canonical home for this skill is issue-triage in cobusgreyling/loop-engineering

SKILL.md
Quality
Evals
Security

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 tight, self-contained instruction skill: every section drives behavior, the output format is fully specified, and safety checkpoints (propose-only L1, escalation, injection handling) are woven in. Remaining gaps are minor — an undefined "loop score", no explicit step ordering, and no label-set verification.

DimensionReasoningScore

Conciseness

The body is lean and assumes competence: a compact output template, a signal-to-priority table, and terse rules with no explanations of concepts Claude already knows. The untrusted-input block is operational guidance that earns its tokens, matching the 5 anchor rather than 4 (which expects trimmable over-explanation).

5 / 5

Actionability

Concrete, executable guidance throughout: a copy-paste markdown state template, a P0–P3 signal mapping table, exact phrasing for duplicates ("possible duplicate of #NNN"), and an enumerated label allowlist. It falls short of 5 due to minor gaps: no command for listing issues ("or Linear/Jira via MCP if configured" is vague) and "Top 5 (by loop score)" references a score never explicitly defined beyond the P0–P3 table.

4 / 5

Workflow Clarity

The sequence is legible (read prior state → scan → score/dedupe → propose → update state) and real checkpoints exist: needs-human escalation domains, human-confirm for duplicates, flag-and-stop on suspected prompt injection, and run-over-run "N (was M)" diffing with pruning of closed issues. It is not 5 because steps are never explicitly ordered and there is no verification that proposed labels exist in the repository's label set.

4 / 5

Progressive Disclosure

No bundle files exist and the body appropriately references none; the only file mentions are runtime state files, and the inlined generated untrusted-input block is intentionally synced by script. Sections are clear and navigable, but at ~70 lines (just over the simple-skill threshold for an automatic 5) with the generated block inlined, structure is good with minor organization notes, fitting the 4 anchor.

4 / 5

Total

17

/

20

Passed

Description

71%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 specific, concise, third-person description with concrete actions and a useful safety constraint. Its main weakness is the complete absence of "when to use" trigger guidance, which caps completeness and weakens trigger-term coverage.

Suggestions

Add an explicit trigger clause, e.g. "Use when asked to triage or groom the issue backlog, deduplicate issues, or suggest labels and priorities."

Include natural user synonyms such as "triage", "backlog", and "GitHub issues" so the skill triggers on phrasings users actually say.

Mention the timing context (e.g. periodic/background runs on a repository) to further distinguish it from one-off issue-management skills.

DimensionReasoningScore

Specificity

"Scan open issues and discussions. Dedupe, prioritize, and propose labels. Updates issue-triage-state.md." lists multiple concrete actions (scan, dedupe, prioritize, propose labels, update state file) plus an explicit constraint ("never auto-label or close"), comprehensively covering the triage pipeline. It matches the 5 anchor rather than 4 because no meaningful capability in the domain is left uncovered.

5 / 5

Completeness

The "what" is clear and multi-part, but there is no "Use when..." clause or equivalent explicit trigger guidance, which caps completeness at 3 per the judging guidelines. It is not a 4 because the "when" is entirely absent rather than present-but-imprecise.

3 / 5

Trigger Term Quality

Natural terms like "issues", "discussions", "dedupe", "prioritize", and "labels" are present, but common user phrasings such as "triage", "backlog", and "GitHub issues" are missing. Good coverage with a few natural terms absent fits the 4 anchor; the 5 anchor requires synonyms and fuller variations.

4 / 5

Distinctiveness Conflict Risk

"Scan open issues and discussions. Dedupe, prioritize, and propose labels" carves out a distinct issue-triage niche, and the L1 propose-only framing is unusual. It is not a 5 because no trigger phrases distinguish it from closely related queue-management skills (the body itself pairs with a sibling "Daily Triage" skill), leaving minor overlap risk.

4 / 5

Total

16

/

20

Passed

Validation

87%

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

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

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

Warning

referenced_paths_exist

Referenced path issues: 1 missing

Warning

Total

14

/

16

Passed

Repository

cobusgreyling/loop-engineering

Path

starters/issue-triage/.codex/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.