CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

investigating-ci-failures

Investigates a specific CI failure to a verdict: whose fault, which commit, who wrote it, and whether it's fixed. Use for "who broke master", "why did this test fail in CI", "is this failure my PR's fault or everyone's", "is this test flaky or actually broken", "when did this failure start". Works from the engineering_analytics warehouse views (engineering_analytics_ci_failures, engineering_analytics_ci_job_history) plus the CI failure logs. Not for aggregate CI health, cost, or merge bottlenecks (use diagnosing-ci-and-merge-bottlenecks) and not for building saved insights (use turning-engineering-analytics-into-insights).

68

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

—

The risk profile of this skill

SKILL.md
Quality
Evals
Security

Quality

Content

70%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 tightly written, well-sequenced investigative workflow rich in domain specifics and validation checkpoints. Its main weakness is that the copy-ready SQL it repeatedly points to lives in a referenced file that is missing from the bundle, which hurts both actionability and progressive disclosure.

Suggestions

Ship references/investigation-queries.md with the skill (or inline the key queries), since the body depends on queries 1, 2, 5, 7, and 8 by number but none are accessible.

Verify the ./references/ path in the body resolves to a real file; a broken reference undermines the one-level-deep navigation the structure otherwise relies on.

Consider inlining at least the boundary query (query 2) and the shape/fingerprint query (query 1) so the core workflow is executable from SKILL.md alone.

DimensionReasoningScore

Conciseness

Lean and free of basic-concept padding — every section is domain knowledge Claude lacks (warehouse views, fingerprints, merge-queue gate mechanics, Depot cache caveat). A few dense sentences could be trimmed, so it sits just below the fully-lean anchor 5.

4 / 5

Actionability

Names concrete MCP tools and columns, but the 'Copy-ready SQL for every step' is delegated to references/investigation-queries.md (queries 1/2/5/7/8 referenced by number, none inline), and that file is absent — leaving key executable details missing. Not 4 because the gap is not minor; not 2 because substantial concrete guidance remains.

3 / 5

Workflow Clarity

A clearly sequenced investigation flow (triage → fingerprint → shape → boundary → culprit → verify) with explicit validation checkpoints and feedback loops: confidence check before naming anyone, mismatch detection, freshness check via max(created_at), and suspicion of a recent failure-free tail. This is a read-only skill, so the destructive/batch cap does not apply.

5 / 5

Progressive Disclosure

Sections are well-organized and the single external reference is one level deep and clearly signaled, but the referenced file references/investigation-queries.md does not exist in the bundle, so navigation to the bulk of the actionable SQL is broken — a defect beyond the minor organization gaps allowed at anchor 4.

3 / 5

Total

15

/

20

Passed

Description

100%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 a strong, third-person statement that crisply states the skill's verdict-oriented purpose, lists natural trigger phrases, and explicitly distinguishes it from two adjacent skills. It cleanly satisfies the what/when/why-niche requirements.

DimensionReasoningScore

Specificity

Enumerates concrete verdict actions — 'whose fault, which commit, who wrote it, and whether it's fixed' — giving comprehensive, specific coverage rather than the 1-2 actions of a score-4 anchor.

5 / 5

Completeness

Explicitly answers both 'what' (investigates a CI failure to a verdict) and 'when' via a 'Use for …' clause with concrete trigger phrases.

5 / 5

Trigger Term Quality

Quotes natural phrases users actually say — 'who broke master', 'why did this test fail in CI', 'is this test flaky or actually broken', 'when did this failure start' — comprehensive coverage including synonyms.

5 / 5

Distinctiveness Conflict Risk

Defines a clear niche and explicitly fences off siblings ('Not for … use diagnosing-ci-and-merge-bottlenecks', 'not for … use turning-engineering-analytics-into-insights'), minimizing wrong-skill triggering.

5 / 5

Total

20

/

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

relative_links

Relative link issues: 1 missing

Warning

referenced_paths_exist

Referenced path issues: 1 missing

Warning

Total

14

/

16

Passed

Repository

PostHog/posthog

Path

products/engineering_analytics/skills/investigating-ci-failures/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.