CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

smoke-check

Critical-path smoke gate before QA hand-off — runs the automated suite. A failed check means the build is not QA-ready.

Quality

77%

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 ./.claude/skills/smoke-check/SKILL.md
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.

This is a well-engineered orchestrator skill: a lean overview that delegates phase detail to six real, well-scoped reference files with unmistakable read-me signals, explicit gating/validation semantics, and a precise argument grammar. The only notable flaws are the six-fold verbatim repetition of the compaction/resume instruction and a couple of motivational sentences that could be trimmed.

DimensionReasoningScore

Conciseness

The body is lean and assumes competence — config semantics (qa.level, automation modes) and project-specific rules are things Claude would not know, so they earn their tokens. Minor trimmable instances keep it below 5: the instruction "Read it again if the conversation was compacted or resumed during this phase" is repeated verbatim in all six phase sections (~90 tokens that could be stated once globally), and "Handing a broken build to QA wastes their time and demoralises the team" is motivational padding.

4 / 5

Actionability

Guidance is concrete and directive: exact argument grammar with defaults ("sprint | quick | --platform pc|console|mobile|all"), exact output path ("production/qa/smoke-[date].md"), exact config keys to resolve, and imperative per-phase pointers ("Read references/1-detect-setup.md now and follow it") to verified, substantive files. It is below 5 because the body itself contains few executable commands — the executable detail lives in the reference files — and the platform-check specifics for Phases 4-5 are only sketched.

4 / 5

Workflow Clarity

The orchestration sequence is explicit (Parse Arguments → Phases 1-6 → gate) with explicit validation semantics and outcome taxonomy ("a build that fails smoke check does not go to QA", "Unconfirmed NOT RUN yields NOT ASSESSED, not FAIL", "PASS WITH WARNINGS does not block"), a feedback loop for error recovery ("quick — use for rapid re-checks after fixing a specific failure"), and a checklist-style Collaborative Protocol with a hard checkpoint ("Never write the report without asking"). Despite batch verification of critical paths, validation is present at every decision point, so the batch cap at 3 does not apply; this matches the anchor-5 pattern of validate → fix → re-check → only-when-valid proceed.

5 / 5

Progressive Disclosure

Scored against the actual bundle: six reference files exist, are substantive (40-233 lines each), map one-to-one onto the six numbered phases, and contain no nested references/ pointers (verified), so references are exactly one level deep. Signaling is unambiguous (bold, imperative "Read references/N-x.md now and follow it"), the SKILL.md body is a ~90-line overview with details appropriately split out, and each reference self-identifies as part of the skill. This matches 'Clear overview with well-signaled one-level-deep references; content appropriately split; easy navigation'.

5 / 5

Total

18

/

20

Passed

Description

62%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 concise, in third person, and clearly communicates the skill's gate role and consequence semantics, with an explicit when-clause ("before QA hand-off"). Its weaknesses are thin action coverage and sparse natural trigger terms — it names the domain well but undersells what the skill actually does and how users would phrase the request.

Suggestions

Add the body's other concrete actions to the description, e.g. "Runs the automated test suite, checks coverage gaps, batch-verifies critical paths, and produces a PASS/FAIL report at production/qa/smoke-[date].md."

Add explicit trigger phrasing: "Use when a build is ready for QA hand-off, or when the user asks for a smoke test, smoke check, or pre-QA test run."

Include natural synonyms users would say — "smoke test", "run the tests before QA", "QA-ready check" — alongside the existing jargon terms like "smoke gate" and "automated suite".

DimensionReasoningScore

Specificity

"runs the automated suite" is one concrete action and "Critical-path smoke gate before QA hand-off" names the domain clearly, but the description stops there — it omits the other concrete capabilities the body lists (coverage-gap checks, batch verification of critical paths, PASS/FAIL report generation). This matches the anchor 'Names domain and 1-2 concrete actions, but not comprehensive'; it is above score 2 because the action stated is concrete rather than generic, and below 4 because there are not several specific actions.

3 / 5

Completeness

Both parts are explicitly stated: the what ("runs the automated suite" plus the verdict semantics "A failed check means the build is not QA-ready") and the when ("before QA hand-off", an explicit trigger situation equivalent to a Use-when clause, so the completeness cap of 3 does not apply). It is below 5 because there are no concrete trigger phrases (e.g. "Use when the user asks for a smoke test before QA hand-off") and the 'what' is minimal compared to the anchor-5 example.

4 / 5

Trigger Term Quality

Relevant keywords are present ("smoke gate", "QA hand-off", "automated suite", "critical-path") but common natural variations users would say — "smoke test", "run the tests", "test suite", "QA-ready" — are missing, and "automated suite" is not the phrase a user would naturally utter. Fits 'Some relevant keywords but missing common variations or synonyms'; not 4 because keyword coverage is thin rather than good-with-few-gaps.

3 / 5

Distinctiveness Conflict Risk

"Critical-path smoke gate before QA hand-off" carves a clear niche (pre-QA gating in a game-dev pipeline) that is mostly distinct from a generic run-the-tests skill, matching 'Mostly distinct; minor overlap risk with closely related skills'. Not 5 because the only distinguishing trigger term is "smoke", so a user asking to 'run the tests' could plausibly hit a competing test-execution skill.

4 / 5

Total

14

/

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

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

frontmatter_unknown_keys

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

Warning

Total

14

/

16

Passed

Repository

Donchitos/Claude-Code-Game-Studios

Path

.claude/skills/smoke-check/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.