CtrlK
BlogDocsLog inGet started
Tessl Logo

ci-triage

Classify CI failures — distinguish clear regressions from infra flakes and security-test failures. Produces structured failure reports.

59

Quality

69%

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

Fix and improve this skill with Tessl

tessl review fix ./starters/ci-sweeper-opencode/skills/ci-triage/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

80%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 lean, well-organized instruction skill with concrete categories and rules, but its auto-fix loop lacks an explicit validation checkpoint before retry, which is the main workflow-clarity weakness.

Suggestions

Add an explicit validation step in the fix loop, e.g. 'After each fix attempt, re-run the failing check; only count the attempt if it was actually re-run.'

Clarify the sequence from classification to action (e.g. classify → decide next action → apply fix in worktree → re-run → update ci-sweeper-state.md) so the workflow is sequenced rather than implied.

Specify what 'retry once before classifying' for non-deterministic failures means concretely (re-trigger the check how, and what to record).

DimensionReasoningScore

Conciseness

The body is lean and assumes Claude's competence — it defines only the classification scheme, output, and rules with no padded explanations of concepts Claude already knows, matching the 'lean and efficient; every token earns its place' anchor.

5 / 5

Actionability

Guidance is concrete: named categories, a specific output file (ci-sweeper-state.md) with explicit fields, and hard rules (max 3 attempts, worktree isolation). It is instruction-level rather than executable code, but for an instruction-only skill that is actionable with only minor gaps, fitting the 4 anchor.

4 / 5

Workflow Clarity

Classification rules and boundaries are clear, but the skill drives batch auto-fix attempts (up to 3 per item) without an explicit validation checkpoint confirming a fix resolved the failure before retrying; the feedback-loop guideline caps such skills at 3.

3 / 5

Progressive Disclosure

At under 50 lines with no need for external references, the well-organized sections (Classification, Output, Rules) satisfy the simple-skill exception, earning the 'clear overview, well-organized' top anchor.

5 / 5

Total

17

/

20

Passed

Description

58%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 specific and well-scoped to a clear niche, but it lacks any explicit 'when to use' trigger guidance, which limits its discoverability and completeness.

Suggestions

Add a 'Use when...' clause naming natural trigger phrases users say, e.g. 'Use when a CI run fails or the user asks why checks are red or whether a failure is a flake.'

Broaden trigger terms with common synonyms such as 'flaky build', 'broken tests', or 'red CI' to improve natural-language matching.

Optionally mention the structured output format (e.g. the ci-sweeper-state.md report) so the 'what' covers the deliverable shape, not just the action.

DimensionReasoningScore

Specificity

Lists several concrete actions — 'Classify CI failures', 'distinguish clear regressions from infra flakes and security-test failures', 'Produces structured failure reports' — with only minor coverage gaps, matching the 'lists several specific actions; minor gaps' anchor rather than the comprehensive 5.

4 / 5

Completeness

The 'what' is clear (classify/distinguish/report), but there is no 'when' guidance at all; per the judging guidelines a missing 'Use when...' clause caps completeness at 3.

3 / 5

Trigger Term Quality

Terms like 'CI failures', 'regressions', 'infra flakes', and 'security-test failures' are relevant but it lacks common user phrasings and synonyms (e.g. 'broken tests', 'flaky build', 'red CI') and has no explicit 'Use when...' trigger, fitting the 'some relevant keywords but missing common variations' anchor.

3 / 5

Distinctiveness Conflict Risk

The CI-triage niche is well-scoped with distinct triggers ('CI failures', 'infra flakes', 'security-test failures') and only minor overlap risk with general debugging/build skills, matching the 'mostly distinct; minor overlap risk' anchor.

4 / 5

Total

14

/

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.

Validation15 / 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
cobusgreyling/loop-engineering
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.