CtrlK
BlogDocsLog inGet started
Tessl Logo

find-skills

Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.

96

4.66x
Quality

83%

Does it follow best practices?

Impact

98%

4.66x

Average score across 11 eval scenarios

SecuritybySnyk

Low

Low-risk findings worth noting

The canonical home for this skill is find-skills in vercel-labs/skills

SKILL.md
Quality
Evals
Security

Evaluation results

100%

10%

Automating Release Notes for a Growing Open Source Project

Can-you-do-X trigger and alternative search terms

Criteria
Baseline
With context

Triggers skill search

100%

100%

Relevant initial keywords

100%

100%

Tries alternative terms

100%

100%

Lists all search attempts

100%

100%

Provides fallback if nothing found

100%

100%

Includes install command if skill found

0%

100%

100%

100%

Explore Available Skills for TypeScript Development

Popular sources awareness and ecosystem discovery

Criteria
Baseline
With context

Executes skill search

0%

100%

Includes domain keyword

0%

100%

Specific compound query

0%

100%

Mentions vercel-labs source

0%

100%

Mentions ComposioHQ source

0%

100%

Presents skills.sh link

0%

100%

Full owner/repo@skill format

0%

100%

100%

60%

Finding a Reliable Database Migration Skill

Quality verification before recommending skills

Criteria
Baseline
With context

Runs skill search

100%

100%

Mentions install count

0%

100%

Prefers high install count

0%

100%

Caution on low installs

0%

100%

Source reputation check

100%

100%

GitHub stars check

0%

100%

Does not recommend blindly

100%

100%

100%

100%

Keep the Skills Ecosystem Current

Skills check and update maintenance

Criteria
Baseline
With context

Runs check command

0%

100%

Runs update command

0%

100%

Documents check results

0%

100%

Documents update results

0%

100%

Correct command names

0%

100%

100%

100%

Respond to a Developer's Frustration

Implicit wish-based trigger recognition

Criteria
Baseline
With context

Recognizes implicit trigger

0%

100%

Executes skill search

0%

100%

Domain keyword in query

0%

100%

Presents skill name and purpose

0%

100%

Provides install command

0%

100%

Includes skills.sh link

0%

100%

100%

96%

Skill Discovery for a Next.js Storefront Project

Leaderboard-first discovery workflow

Criteria
Baseline
With context

Checks leaderboard first

0%

100%

References leaderboard by name

0%

100%

Identifies vercel-labs/agent-skills

0%

100%

Provides correct install command

0%

100%

Documents discovery steps in order

26%

100%

100%

100%

Task

Recognizes 'how do I' trigger and formulates effective search query with domain and task keywords

Criteria
Baseline
With context

Recognizes trigger phrase

0%

100%

Executes search command

0%

100%

Includes domain keyword

0%

100%

Includes task keyword

0%

100%

80%

60%

Task

Presents search results with complete information and installs skills with correct flags

Criteria
Baseline
With context

Presents skill name and purpose

100%

100%

Provides install command

0%

100%

Includes skills.sh link

0%

0%

Uses global flag

0%

100%

Uses auto-confirm flag

0%

100%

100%

50%

Task

Handles empty search results gracefully with acknowledgment, direct help offer, and custom skill suggestion

Criteria
Baseline
With context

Executes search attempt

0%

100%

Acknowledges no results

88%

100%

Offers direct assistance

93%

100%

Suggests creating custom skill

0%

100%

100%

70%

Task

Uses category knowledge to translate vague requests into effective search queries with appropriate domain keywords

Criteria
Baseline
With context

Maps request to category

100%

100%

Uses category keywords

0%

100%

Includes framework context

0%

100%

100%

100%

Task

Recognizes interest in extending capabilities as skill search trigger and presents results with proper format

Criteria
Baseline
With context

Recognizes capability extension trigger

0%

100%

Searches for relevant skills

0%

100%

Presents actionable results

0%

100%

Repository
Ed1s0nZ/CyberStrikeAI
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

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.