CtrlK
BlogDocsLog inGet started
Tessl Logo

browse-and-evaluate

Use when exploring the ai-agent-skills catalog to find, compare, and evaluate skills before installing. Always use --fields to limit output size and --dry-run before committing to an install.

68

Quality

84%

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

93%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 an exemplary lean workflow skill: every step is an executable command with sensible flags, guardrails double as validation gates (dry-run before install, confirmation for batch installs), and the section structure is clean with no bloat or unnecessary explanation. The only meaningful gap is the absence of an explicit recovery path when a dry-run or source check fails — the gotchas hint at the risks but do not close the feedback loop.

DimensionReasoningScore

Conciseness

The body is lean throughout — a goal line, five terse guardrail bullets, five one-line workflow steps each with a single command, and three gotchas — with no explanation of concepts Claude already knows; every token earns its place, matching the 'lean and efficient' anchor. It is not 4 because there is no padded or over-explanatory passage to trim (the one elaborative note, 'The CLI defaults to JSON when stdout is not a TTY', is operationally useful, not filler).

5 / 5

Actionability

Every workflow step is a copy-paste-ready command with concrete flags, e.g. 'npx ai-agent-skills search <query> --fields name,tier,workArea,description --limit 10' and 'npx ai-agent-skills install <skill-name> --dry-run', covering all common cases (search, info, preview, dry-run, install), matching the 'fully executable' anchor. It is not 4 because there are no gaps — placeholders are appropriate parameterization, and specific flag values are given rather than left abstract.

5 / 5

Workflow Clarity

The five-step sequence is clearly ordered with an explicit validation checkpoint — step 4 dry-runs the install and step 5 says 'Install only after reviewing the dry-run output' — plus guardrails like 'Never install more than 3 skills at once without explicit user confirmation', matching 'clear sequence with most checkpoints present'. It is not 5 because there is no explicit error-recovery loop for a failed validation (e.g., what to do when the dry-run output looks wrong or the upstream source is unreachable beyond the one-line gotcha hint).

4 / 5

Progressive Disclosure

The skill is a compact, single-purpose workflow (~50-line body) with no references/, scripts/, or assets/ bundle directories, and no external references are needed — the Goal / Guardrails / Workflow / Gotchas sections are well-organized and fully self-contained, which the rubric's simple-skill guidance allows to score 5. It is not 4 because there is no inline content that should have been split out and no buried or nested references; the structure is exactly at the right altitude for the content's size.

5 / 5

Total

19

/

20

Passed

Description

75%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 solid description that explicitly pairs a clear capability statement with a 'Use when' trigger and stays lean without buzzwords. Its main weakness is moderate trigger coverage — it names the niche well but misses natural synonyms and trigger variations that would push completeness and trigger quality to the top anchor. The behavioral clause about --fields/--dry-run is actionable but does not add to what/when clarity.

Suggestions

Add natural trigger synonyms and variations to the 'when' clause, e.g., 'Use when the user mentions browsing, searching, comparing, or evaluating skills, or asks what to install from the ai-agent-skills catalog.'

Broaden trigger-term coverage with user-natural phrases like 'find the right skill', 'skill discovery', or 'preview a skill before installing' so the description matches how users actually phrase the request.

Consider naming one more concrete capability (e.g., 'preview skill content and check install sources') to move specificity from several-actions to comprehensive coverage.

DimensionReasoningScore

Specificity

The description lists several concrete actions — 'to find, compare, and evaluate skills before installing' — grounded in a specific domain (the ai-agent-skills catalog), matching the 'several specific actions; minor gaps in coverage' anchor. It falls short of 5 because the actions are stated at a high level without additional concrete operations (e.g., previewing skill content or checking install sources), and it sits above 3 because more than 1-2 specific actions are named.

4 / 5

Completeness

It has both a clear 'what' ('find, compare, and evaluate skills') and an explicit 'when' ('Use when exploring the ai-agent-skills catalog ... before installing'), matching the 'both present; when could be more explicit' anchor. It does not reach 5 because the 'when' clause lacks concrete trigger variations (e.g., 'when the user mentions skills, catalog, or installing a skill'), and the second sentence ('Always use --fields ... --dry-run ...') is behavioral guidance rather than what/when clarification.

4 / 5

Trigger Term Quality

Natural keywords like 'exploring the ai-agent-skills catalog', 'find', 'compare', 'evaluate skills', and 'installing' give good coverage of phrases a user would say, matching the 'good keyword coverage; a few natural terms missing' anchor. It is not 5 because common variations such as 'browse skills', 'search the catalog', 'skill discovery', or 'preview a skill' are absent, and not 3 because the included terms are specific and user-natural rather than generic.

4 / 5

Distinctiveness Conflict Risk

The trigger is scoped to a named niche ('the ai-agent-skills catalog') with distinct triggers around browsing and installing, matching 'mostly distinct; minor overlap risk'. It is not 5 because generic fragments like 'evaluate skills' could overlap with unrelated skill-evaluation or skill-authoring skills, and not 3 because the catalog-specific framing makes mis-triggering unlikely.

4 / 5

Total

16

/

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
MoizIbnYousaf/ai-agent-skills
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.