Dev & Engineering skill-discoveryskill-installationcliagent-ecosystempackage-managergithub

Skill Discovery Assistant

Helps users discover, evaluate, and install suitable agent skills from the open ecosystem.

FollowSkills review · FSRS-2.0
Not recommended
49/ 100 5-point scale 2.5 / 5
1 2 3 4 5 6
1Trust12 / 25 · 2.4/5

The skill instructs users to install third-party skills via npx skills and suggests using -y to skip confirmation, increasing supply-chain risk, but the skill itself performs no malicious actions and requests no additional permissions. Deductions: recommendation of -y skipping confirmation, lack of strong source verification (e.g., signatures), and insufficient data-flow transparency.

2Reliability7 / 20 · 1.8/5

The skill provides clear steps but relies on external CLI and network services; static review cannot verify actual execution. Deductions: no error handling or failure feedback, handling of abnormal inputs (e.g., invalid queries) unclear.

3Adaptability9 / 15 · 3.0/5

The skill identifies multiple user query scenarios with fairly clear trigger conditions but lacks explicit non-fit boundaries (when not to use). Deductions: potential false triggering for unusual skills or outdated info; environment fit (e.g., mainland China network) not addressed.

4Convention9 / 15 · 3.0/5

The skill documentation is well-structured with name, description, usage scenarios, steps, and examples, but lacks version history, changelog, or maintainer information. Deductions: missing clear maintenance responsibility and change records.

5Effectiveness7 / 15 · 2.3/5

The skill theoretically helps users discover and install skills, but no evidence of actual effectiveness is provided. Deductions: output format (e.g., search results) not explicitly standardized, and dependence on external service availability.

6Verifiability5 / 10 · 2.5/5

The repository includes test files (e.g., archive.test.ts), but these are low-level functionality tests not directly covering this skill's behavior. Deductions: lack of reproducible tests or third-party verification evidence for the skill itself.

Evidence confidence:Low Reviewed Aug 07, 2026 Reviewed revision a4d243c3d4f8
The upstream repository has new commits since this review. The score still applies to the reviewed revision shown and may not cover the latest changes.
Before you use it
  • The skill recommends using -y to skip confirmation, which may lead users to install risky or malicious skills unknowingly; it is advisable to disallow this option when trust is insufficient.
  • Before installing third-party skills, thoroughly verify their source and security, especially in mainland China network environments where external services (e.g., skills.sh, GitHub) may be unreachable; consider alternatives.
  • The skill lacks troubleshooting or error-handling guidance; it is recommended that maintainers supplement and test abnormal scenarios before actual use.
Review evidence [1][2][3][4][5][6][7][8][9]
See the full review method →

What does this skill do, and when should you use it?

This skill assists users in finding and installing skills from the open agent skills ecosystem when they need specific functionality. It guides users to clarify their needs, checks the skills.sh leaderboard for popular options, and runs `npx skills find` to search for relevant skills. Before recommending, it verifies skill quality based on install counts, source reputation, and GitHub stars. Once a suitable skill is found, it provides installation commands and details, and can install directly. If no skill matches, it suggests using general capabilities or creating a custom skill.

Handles queries like 'how do I do X', 'find a skill for X', or 'is there a skill that can...'. It performs: identifying domain and specific task; checking skills.sh leaderboard; running npx skills find [query] [--owner <owner>]; verifying quality via install count (prefer 1K+), source reputation, and GitHub stars; presenting options with description, install count, source, and install command; installing via npx skills add <owner/repo@skill> -g -y if requested; suggesting direct help or npx skills init when no match.

  1. A developer asks how to optimize React performance and needs a relevant skill.
  2. A user wants to perform PR reviews and wonders if a skill exists.
  3. A user needs to generate a changelog and asks for a skill.
  4. A user wants to extend their agent's capabilities but doesn't know where to start.
  5. A user wants to find skills for testing or deployment automation.

What are this skill's strengths and limitations?

Pros
  • Provides clear search steps and validation criteria to avoid low-quality recommendations.
  • Supports both interactive and keyword-based searches.
  • Integrates installation flow, enabling direct install from search results.
  • Offers common skill categories and search tips, improving efficiency.
  • Provides fallback options and guidance for creating custom skills when no match is found.
Limitations
  • Relies on network access and skills.sh; unusable offline.
  • Search and validation depend on external ecosystem, quality may vary.
  • Requires CLI usage, potentially steep for non-technical users.
  • May miss niche or newly emerging skills.

How do you install this skill?

This skill is part of the vercel-labs/skills repository. To install the full skill package, run: npx skills add vercel-labs/agent-skills. For standalone installation, copy the skills/find-skills directory into your agent's skills folder, ensuring SKILL.md is present.

How do you use this skill?

Simply ask your agent, e.g., 'Help me find a skill for React performance' or 'Is there a skill for PR review?' The agent will trigger this skill and follow the search and recommendation process. You can also manually run npx skills find react performance or browse skills.sh.

FAQ

Do I need to pay to use this skill?
No, this skill is free and part of the open ecosystem. However, some third-party skills or tools may have associated costs.
Will installing skills slow down my agent?
Installing skills only adds instruction files and does not directly affect performance. At runtime, skills may invoke external tools and consume resources, depending on their implementation.
What if no skill matches my request?
The skill will suggest using your agent's general capabilities, or guide you to create your own skill via `npx skills init`.
What should I do if skill installation fails?
Check network connectivity, verify the skill name and source, and ensure you have write permissions. You can also manually copy the skill directory to the correct path.

Related skills