find-skills

Run this BEFORE any package install (pip / npm / apt / brew / cargo / gem / go install) you would otherwise execute via the exec tool — including when the user asks for a deliverable that needs third-party libraries (PDF generation, image processing, scrapers, dashboards, charts, OCR, audio/video, …). It searches the open agent-skill ecosystem (skills.sh) for an installable skill that already does the job; only fall back to direct package install if find-skills returns no credible match. Also covers explicit "find a skill for X" / "how do I do X" / "is there a skill that can..." requests, and any time the user expresses interest in extending agent capabilities.

Quality

87%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

85%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable, well-sequenced skill body with concrete commands and a clear feedback loop. Its main weakness is conciseness: the install-count guidance and anti-reimplementation rationale are each repeated multiple times.

Suggestions

Consolidate the install-count guidance into one place (Step 4) and have Step 5 reference it rather than restating 'the 20-install case is FINE'.

Merge the 'don't reimplement' reasoning so it appears once (Step 7) instead of being restated in Step 4 and the anti-pattern section.

The 'Common Skill Categories' table and 'Tips for Effective Searches' overlap with Step 3's examples — consider trimming to reduce token load.

Dimension	Reasoning	Score
Conciseness	Mostly operational and free of concepts Claude already knows, but the install-count reassurance is restated across Step 4 ('Install counts in the high tens / low hundreds are FINE') and Step 5 ('The 20-install case is FINE'), and the 'don't reimplement' rationale is repeated in Step 4, Step 7, and the anti-pattern — tightening these would remove padding.	2 / 3
Actionability	Fully executable, copy-paste-ready commands throughout: `npx skills find <query> 2>&1 \| sed -E 's/\x1b\[[0-9;]*[a-zA-Z]//g'`, `npx skills add <owner/repo@skill> -g -y`, and `ls ~/.agents/skills/<name>/`, with concrete worked examples per query type.	3 / 3
Workflow Clarity	A clearly sequenced Step 1–7 workflow with explicit checkpoints (skim SKILL.md first line, verify URL resolves, reject mismatches) and a feedback loop ('If the skill doesn't fit after reading its docs, uninstall it and try a different one').	3 / 3
Progressive Disclosure	No bundle files are provided, so all content lives in one file; it is well-organized into clearly signaled sections (When to Use, Steps 1–7, Categories, Tips, No Skills Found) with no nested references and easy header-based navigation.	3 / 3
	Total	11 / 12 Passed

Description

90%

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 high-quality description with explicit triggers, natural keyword coverage, and a clear distinctive niche. The only issue is second-person voice ('you would otherwise execute'), which the rubric penalizes on specificity.

Suggestions

Rewrite in third person to avoid the voice penalty: e.g. 'Run BEFORE any package install that would otherwise be executed via the exec tool' instead of 'you would otherwise execute'.

Consider trimming the parenthetical deliverable list slightly so the core trigger ('BEFORE any package install') stays the most prominent signal.

Dimension	Reasoning	Score
Specificity	Lists concrete actions ('searches the open agent-skill ecosystem (skills.sh) for an installable skill', 'fall back to direct package install'), which would normally anchor at 3, but the second-person phrasing ('you would otherwise execute via the exec tool') triggers the voice penalty, reducing specificity by 1.	2 / 3
Completeness	Explicitly answers both 'what' (searches skills.sh for an installable skill, falls back to direct install) and 'when' ('Run this BEFORE any package install', 'Also covers explicit "find a skill for X" / "how do I do X"...'), with explicit trigger guidance rather than implied.	3 / 3
Trigger Term Quality	Strong coverage of natural user phrasings — 'find a skill for X', 'how do I do X', 'is there a skill that can...', plus concrete package managers (pip/npm/apt/brew/cargo/gem/go) and deliverable types (PDF generation, OCR, dashboards) users actually name.	3 / 3
Distinctiveness Conflict Risk	Clear niche — intercepting package installs to find a prebuilt skill — with distinct triggers tied to install intent and 'find a skill' requests, making overlap with other skills unlikely.	3 / 3
	Total	11 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning

	Total	14 / 16 Passed

Repository: fastclaw-ai/fastclaw
Path: internal/agent/bundled_skills/find-skills/SKILL.md
Commit: 533a138

Reviewed: about 17 hours 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.