An exemplary lean, action-first SKILL.md: executable commands with worked examples, complete flag documentation, explicit output expectations, and troubleshooting notes that all convey information Claude would not otherwise know. The only notable weakness is bundle integrity — the referenced search.mjs is not present in the bundle, leaving the body's single reference unverifiable.

Suggestions

Ship search.mjs in the bundle's scripts/ directory (and point the usage examples at scripts/search.mjs) so the body's only file reference is verifiable.

Document the default provider and a concrete --model example value so users don't have to guess valid model IDs.

Briefly state what --json emits so the machine-readable output path is as actionable as the default one.

Dimension	Reasoning	Score
Conciseness	Every section (Script, Usage, Optional flags, Output expectations, Notes) carries non-obvious operational information — env-var fallbacks, module-path overrides, cached-token fallback — with zero padding or explanation of concepts Claude already knows. This matches the 'lean and efficient; every token earns its place' anchor.	5 / 5
Actionability	The usage section gives copy-paste ready commands with two concrete examples ("latest python release", "vite 7 breaking changes") plus a purpose flag, and enumerates all optional flags with their value formats. The common case is fully covered, matching the 'fully executable; specific examples cover the common cases' anchor.	5 / 5
Workflow Clarity	This is a simple, single-purpose skill (run one script, read its output) and the single action is unambiguous: the command, its arguments, and the expected output shape are all specified. Per the rubric's simple-skill guidance, workflow clarity scores 5; there are no destructive or batch operations that would require validation checkpoints.	5 / 5
Progressive Disclosure	The body is well-organized into clear sections with a single, clearly signaled one-level-deep reference ("## Script — search.mjs") that appropriately keeps the implementation out of SKILL.md. It falls short of 5 because the referenced script file is not present in the skill bundle (no scripts/ directory), so the primary reference cannot be verified against the actual bundle structure.	4 / 5
	Total	19 / 20 Passed

Description

78%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 concise, third-person description that clearly states both what the skill does and when to use it, with reasonable natural trigger phrases. Its main gaps are limited action enumeration (a single action plus deliverables) and missing common synonym phrasings like "search the web" or "look up online".

Suggestions

Enumerate the concrete actions more fully (e.g., "Runs a web search on a topic, returns a concise summary keyed to a stated purpose, and lists full canonical URLs with disagreements between sources") to raise specificity.

Add natural trigger variations users would actually say — "search the web", "look this up online", "find the latest X" — to improve trigger term coverage.

Sharpen distinctiveness by contrasting scope, e.g., "for quick lookups, not multi-source deep research" to reduce overlap with research skills.

Dimension	Reasoning	Score
Specificity	The description names one concrete action ("Trigger native web search") plus two output deliverables ("concise summaries and full source URLs"), which matches the anchor 'names domain and 1-2 concrete actions, but not comprehensive'. It is not a 4 because it does not list several specific actions, and not a 2 because the actions go beyond a generic domain label.	3 / 5
Completeness	It explicitly answers both: what ("Trigger native web search... concise summaries and full source URLs") and when ("Use when you need quick internet research with concise summaries and full source URLs"), with a concrete 'Use when...' trigger clause. Not a 4, since the 'when' is explicit rather than merely present-but-vague.	5 / 5
Trigger Term Quality	Phrases like "native web search", "quick internet research", "concise summaries", and "full source URLs" give good natural keyword coverage a user would plausibly say. It falls short of 5 because common variations such as "search the web", "look this up online", or "latest news" are missing.	4 / 5
Distinctiveness Conflict Risk	"Native web search" carves out a mostly distinct niche (fast, native-tooling search with URL-backed summaries), but "web search" / "internet research" is a broad trigger space with minor overlap risk against general research or deep-research skills. Not a 5 because the overlap with other research-oriented skills is more than minimal.	4 / 5
	Total	16 / 20 Passed

Validation

100%

Warnings & errors only

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: mitsuhiko/agent-stuff
Path: skills/native-web-search/SKILL.md
Commit: 0865c84

Reviewed: 8 days ago

Table of Contents

Discovery Implementation Validation

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.