CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

real-world-qa

Tests a change the way a user meets it, in real projects or the real tool, and compares the result on the base and on this head. Use when the user asks for real-world QA, to try a change in real projects, or to check that it works outside the tests.

Quality

91%

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

100%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, highly actionable instruction-only skill: concrete commands and file paths throughout, explicit per-step exit criteria with a finding/fix feedback loop, and a genuinely useful security section for low-trust sources. No dimension shows more than negligible room for improvement.

DimensionReasoningScore

Conciseness

Every line carries operational weight ("Install without lifecycle scripts (`npm install --ignore-scripts`, or `YARN_ENABLE_SCRIPTS=false` for Yarn)") with no explanation of concepts Claude already knows. The three-sentence intro orients rather than pads, so it does not fall to the 'minor instances that could be trimmed' anchor.

5 / 5

Actionability

Commands are copy-paste ready ("gh search code", "yarn task sandbox --template <key>", "cd code && yarn local-registry --help", "npm install --ignore-scripts") and specific examples cover the common cases, with placeholders deliberately routed to named references. No step lacks a concrete command or a concrete artifact to inspect, so it exceeds the 'minor gaps' anchor.

5 / 5

Workflow Clarity

Four sequenced steps each with explicit exit criteria ("Done when you can say what to observe and what the expected result is"; "Done when each target has a base result and a head result"), an explicit finding rule ("A difference you did not expect is a finding"), and a fix-or-report feedback loop ("Fix the findings, or give them to the user"). The destructive/batch cap does not apply because the skill mandates staying read-only toward target projects.

5 / 5

Progressive Disclosure

At roughly 35 well-organized lines ("## Steps", "## Low-trust sources") with clearly signaled, one-level-deep references ([Sandboxes](../../guidelines/sandboxes.md), the `canary` skill), the under-50-lines exception applies. Content is appropriately split: the workflow stays inline while sandbox and canary details live in their own referenced locations.

5 / 5

Total

20

/

20

Passed

Description

83%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: it states a precise, concrete capability and pairs it with an explicit "Use when" clause containing natural trigger phrases. The remaining room is in covering the reporting half of the skill and broadening trigger synonyms.

Suggestions

Extend the 'what' clause to cover the skill's output, e.g. '...and reports the findings with base-vs-head evidence', so the capability list is comprehensive and specificity can reach 5.

Add a natural synonym trigger such as 'manually test a change' or 'verify a change in the wild' to broaden keyword coverage.

Tie the 'check that it works' trigger more tightly to the base-vs-head comparison (e.g. 'check that it works outside the tests, compared against the base') to reduce overlap with generic test-running skills.

DimensionReasoningScore

Specificity

Names concrete actions — "Tests a change... in real projects or the real tool", "compares the result on the base and on this head" — with manner and venue precision beyond the generic 1-2-action anchor. Not 5 because it omits part of the skill's scope, such as reporting and fixing findings.

4 / 5

Completeness

Explicitly answers both what ("Tests a change the way a user meets it... and compares the result on the base and on this head") and when ("Use when the user asks for real-world QA, to try a change in real projects, or to check that it works outside the tests") with three concrete trigger phrases — the top anchor.

5 / 5

Trigger Term Quality

"real-world QA", "try a change in real projects", and "check that it works outside the tests" are phrases a user would naturally say when they need this skill. Not 5 because common synonyms like "manually test" or "in the wild" are missing.

4 / 5

Distinctiveness Conflict Risk

"real-world QA", "real projects", and "base and... head" carve a distinct niche with minimal overlap. Not 5 because the trigger "check that it works outside the tests" could also fire for a generic run-the-tests request.

4 / 5

Total

17

/

20

Passed

Validation

75%

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

Validation — 12 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

relative_links

Relative link issues: 2 suspicious

Warning

referenced_paths_exist

Referenced path issues: 1 missing

Warning

Total

12

/

16

Passed

Repository

storybookjs/storybook

Path

.agents/skills/real-world-qa/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.