Agent Skill Optimization — the new SEO

Make sure coding agents
use your skill

Developers don't Google for tools anymore — they ask their AI agent. The agent reads skills, picks one, and executes. Skills are the new SEO.

GTM has changed

Business decisions now happen inside the terminal. The AI agent is the new buyer — and it's choosing between your skill and your competitor's right now.

$ Developer: "Find the latest AI news and summarize the key points"
$ Agent evaluating available skills...
Your Search Skillselected
Competitor's Web Scraper
Built-in WebSearch
One skill wins the context window. The rest are invisible.

Built for anyone with a skill

If you've built a skill, plugin, or MCP tool for AI agents, you need to know if agents are choosing it.

API companies

Your skill teaches agents to call your API. Make sure they actually do — not your competitor's.

SaaS platforms

Your skill teaches agents to interact with your product. But is it winning against alternatives?

Developer tool makers

Your CLI or SDK skill competes with others. A better-worded description can steal your traffic.

Open-source projects

Your skill helps agents recommend and use your library. See if it gets picked over the alternatives.

Before & after

"I wrote a skill, I hope agents use it"

"I know my selection rate is 87% and improving"

"No idea if competitors are stealing my traffic"

"I caught 3 steals and fixed my description to prevent them"

"Every agent might choose differently and I can't test"

"I tested across 4 agents and optimized for each"

"My skill description is just... what I wrote"

"My skill description is A+ graded and battle-tested"

How it works

Three steps to know if your skill wins

1

Submit your skill

Paste your skill description, upload a .md file, or search the skills.sh repository.

2

Battle the competition

Your skill enters the arena against real competitors. AI agents simulate choosing between them.

3

See the verdict

Get your selection rate, detailed reasoning for each decision, and insights to improve.

LIVE BATTLE RESULT
WINNER
Your Search Skill

"Advanced web search with real-time results and structured output..."

72%
selection rate
VS
Competitor Skill

"Basic web scraping tool for fetching page contents..."

28%
selection rate
Agent's reasoning

"The winning skill clearly specified real-time capabilities and structured output, which directly matched the task of finding current news. The competitor only mentioned basic page fetching without real-time or formatting guarantees."

Everything you need

Benchmark, compare, and optimize your agent skills

Head-to-head battles

Compare your skill directly against competitors in realistic agent scenarios.

Selection rate metrics

See exactly how often AI agents choose your skill — with clear percentages.

Battle royale & ELO

Pit 3+ skills against each other. Get ELO rankings to see where you stand in your category.

Actionable insights

AI-powered reasoning explains why your skill was chosen — or wasn't.

Open source SDK

Run evaluations locally or in CI/CD. pip install skills-arena.

Multi-agent testing

Test across Claude, GPT, and more to see how different agents evaluate your skill.

Or use the SDK

Run evaluations locally, in CI/CD, or anywhere Python runs.

$ pip install skills-arena
>>> from skills_arena import Arena
>>> results = Arena().evaluate("./my-skill.md", task="web search")
>>> print(results.selection_rate)
0.72

"You spent months building a great product. You wrote a skill so agents can use it. But right now, inside thousands of terminals, an AI agent is reading your skill description next to your competitor's — and choosing theirs."

Make sure they use yours

Two free evaluations — no signup needed.

Enter the Arena