CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

insight-stand

Operate the Insight compose stand and run the deployed-stand suite in tests/stand/ — bring it up (`./dev-compose.sh test-stand up`), seed it, aim a run at it (local or remote), read what it was seeded with, and triage the failures that are the stand rather than the product. Use whenever a task means running, re-seeding, pointing, or debugging the stand: 'bring the stand up', 'run the stand tests', 'test-stand up failed', 'why did collection abort', 'UsageError: stand manifest unusable', 'these tests all skipped', 'the login loops', 'scratch rows survived the run', 'point the suite at another stand', 'what personas does the stand have'. This is the environment skill; stand-api-test and stand-ui-test own the test code, stand-scenarios owns what to test, and drive-ui owns driving a browser at a stand by hand.

Quality

94%

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

SKILL.md
Quality
Evals
Security

Quality

Content

88%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 strong, high-signal operations skill: every command is copy-paste executable, the workflow carries real validation checkpoints and an error-recovery triage table, and the destructive reset is explicitly guarded. The only slack is minor rhetorical padding and reference-shaped tables inlined in an already long single file.

Suggestions

Trim rhetorical framing sentences (e.g. around "the one rule that shapes everything" and "gets diagnosed as flakiness") to pure statement-of-fact, saving tokens without losing guidance.

Move the session-fixture catalogue and base-URL trap detail into a reference file under references/ and link them from a short section, shrinking SKILL.md toward a lean overview.

Consider a compact "old patterns / deprecated" note for time-bound statements ("no suite image is published anymore", the Chromium 149 measurement) so future readers can spot staleness quickly.

DimensionReasoningScore

Conciseness

The body is dense and almost entirely project-specific (instance flags, base-URL precedence chains, netns constraint) with no generic-concept padding, matching the 4 anchor. Not 5: a few sentences are rhetorical rather than load-bearing ("That is the kind of failure that gets diagnosed as flakiness", the discursive "one rule that shapes everything" framing); not 3 since no whole section is unnecessary.

4 / 5

Actionability

Copy-paste-ready commands throughout: the dev-compose verb block, the uv invocations with full env-var exports for aiming a run, the direct pytest command with --base-url/--stand-manifest, and the coverage-gate invocation, plus concrete marker/fixture/triage tables. Fully executable and covering the common cases — the 5 anchor.

5 / 5

Workflow Clarity

Clear up→seed→test→down sequence with explicit checkpoints ("block until dbt built gold", PROFILE.md `--check`, the coverage gate, the `no_scratch_rows_survive` leak check) and a triage table that is a literal symptom→cause→fix feedback loop. The destructive `down` (volume teardown) is guarded by the mandatory `--instance` rule with an explicit warning that omitting it deletes a developer's databases, so the destructive-operation cap does not apply. 5 rather than 4 because validation and recovery are explicit, not implicit.

5 / 5

Progressive Disclosure

Well-organized single-file skill with clear section headers, tables, and a clearly signaled external reference (PROFILE.md, with its regeneration and --check commands). No bundle files exist, so this scores against structure alone: 4 rather than 5 because at ~290 lines reference-shaped material (the eight-row session-fixture catalogue, the base-URL trap detail) is inlined that could live in a reference file; not 3 because the content is cohesive operations knowledge rather than bulk API reference, and navigation is easy.

4 / 5

Total

18

/

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.

An exemplary description: concrete third-person action list with the real command, an explicit use-when clause, ten natural trigger phrases including verbatim error strings, and explicit boundary statements against sibling skills. It answers what, when, and which-skill in one compact paragraph.

DimensionReasoningScore

Specificity

The description lists five concrete actions — "bring it up (`./dev-compose.sh test-stand up`), seed it, aim a run at it (local or remote), read what it was seeded with, and triage the failures" — including the actual command, in third person. This matches the 5 anchor (multiple specific concrete actions, comprehensive coverage); a 4 would require minor coverage gaps, and the actions span the skill's full operate/seed/aim/read/triage scope.

5 / 5

Completeness

Both halves are explicit: "what" is the concrete operate/seed/aim/read/triage action list, and "when" is "Use whenever a task means running, re-seeding, pointing, or debugging the stand" followed by concrete trigger phrases. This is the 5 anchor verbatim in structure; a 4 would mean the "when" clause was less explicit, which it is not.

5 / 5

Trigger Term Quality

Ten quoted natural trigger phrases including exact error strings users would paste ("test-stand up failed", "UsageError: stand manifest unusable", "the login loops", "scratch rows survived the run", "what personas does the stand have"). Comprehensive coverage including synonyms and verbatim failure messages; not a 4 since common variations are already represented.

5 / 5

Distinctiveness Conflict Risk

It carves out a clear niche ("This is the environment skill") and explicitly disambiguates four sibling skills ("stand-api-test and stand-ui-test own the test code, stand-scenarios owns what to test, and drive-ui owns driving a browser at a stand by hand"), with distinct error-message triggers. Minimal conflict risk; the 5 anchor fits.

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

allowed_tools_field

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

Warning

relative_links

Relative link issues: 1 suspicious

Warning

Total

14

/

16

Passed

Repository

constructorfabric/insight

Path

.claude/skills/insight-stand/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.