CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

triage

Triage a GitHub issue or a reported problem — check it is real and not by design (code, git history, PRs, Hindsight memory), find the code and patterns to reuse, build a proof-of-concept fix and prove it solves the problem, then explain the problem and the fix in behaviours before doing the full fix. Use when asked to "triage

Quality

94%

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

96%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 excellent instruction-only workflow skill: lean token-efficient prose, copy-paste-ready commands, a five-step sequence with explicit validation checkpoints and an early-stop path, and clean section structure. The only structural improvement would be splitting the specialized perf and UI-service guidance into reference files.

DimensionReasoningScore

Conciseness

Dense, imperative prose with zero explanation of concepts Claude already knows — every paragraph carries repo-specific policy or concrete commands (e.g., the entire "No workarounds" paragraph enumerates specific prohibited anti-patterns rather than generic advice). No passage can be trimmed without losing an instruction, matching the 5 anchor rather than the 4 anchor's "minor instances of over-explanation".

5 / 5

Actionability

Fully executable guidance throughout: copy-paste-ready commands like `gh issue view <N> --comments`, `git log -S '<key string>' --oneline`, `lsof -nP -iTCP:<port> -sTCP:LISTEN`, `HINDSIGHT_API_DATABASE_URL=pg0://triage-<issue-number-or-short-slug>`, `pg0 drop --name <name> -f`, plus concrete branch naming and perf tool paths. For an instruction-only skill this meets the 5 anchor's copy-paste-ready standard with only issue-specific placeholders, which are inherently variable.

5 / 5

Workflow Clarity

Five numbered, clearly headed steps with explicit validation checkpoints: "The reproduction from step 1 fails before the change and passes after it. Nearby tests still pass", an early-stop checkpoint in step 2 ("If it's not real... tell the user plainly why, with the evidence"), and a final user-decision gate ("Don't start it before they answer"). The reproduce → POC → re-verify-other-paths loop matches the 5 anchor's explicit validation and feedback loops; not the 4 anchor, since no checkpoint is missing.

5 / 5

Progressive Disclosure

Well-organized single file with clear section headers and one clearly signaled one-level-deep reference ("Read `.claude/skills/code-review/SKILL.md`"). It falls short of the 5 anchor because the ~95-line body inlines specialized perf-benchmark and UI-services guidance that could live in separate reference files, and it exceeds the under-50-lines case where section headers alone would suffice; it is comfortably above the 3 anchor's disorganized-inline-content example.

4 / 5

Total

19

/

20

Passed

Description

92%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: third-person voice, comprehensive concrete capabilities, and an explicit 'Use when' clause with three quoted natural trigger phrases. The only weakness is slightly incomplete synonym coverage in its trigger terms.

DimensionReasoningScore

Specificity

Lists multiple concrete actions covering the whole workflow end-to-end: "check it is real and not by design (code, git history, PRs, Hindsight memory)", "find the code and patterns to reuse", "build a proof-of-concept fix and prove it solves the problem", "explain the problem and the fix in behaviours". Coverage is comprehensive with no minor gaps, matching the 5 anchor rather than the 4 anchor.

5 / 5

Completeness

Explicitly answers both questions: a detailed multi-action "what" and "Use when asked to 'triage #N', 'look into this issue', or 'is this a real bug'" with concrete trigger phrases. This mirrors the 5 anchor example's structure exactly; the 4 anchor's "'when' could be more explicit" caveat does not apply.

5 / 5

Trigger Term Quality

Explicitly quotes natural user phrases — "triage #N", "look into this issue", "is this a real bug" — plus domain terms (GitHub issue, bug, problem), which is well above the 3 anchor's partial keyword coverage. A few natural synonyms like "investigate", "reproduce", or "check this issue" are missing, so it falls short of the 5 anchor's comprehensive synonym coverage.

4 / 5

Distinctiveness Conflict Risk

Occupies a clear niche — verifying a reported issue before committing to a fix — with distinct triggers ("triage #N", "is this a real bug") unlikely to fire a general fix or code-review skill. Not the 4 anchor, because the overlap risk with closely related skills is minimal.

5 / 5

Total

19

/

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

vectorize-io/hindsight

Path

.claude/skills/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.