Agent Skills Radar · Finder

Find AI agent skills — what each one actually does

Search and filter a curated directory of AI agent skills, harnesses, and toolkits on GitHub. Short plain-English blurbs, live star signals, and honest links — no README paste, no ownership claims.

Updated 2026-09-20 · 10 curated projects · daily star refresh

Get the daily top 5

Digest waitlist — plain-English picks of the hottest AI agent skills and harnesses. Not a newsletter blast dump; a short daily shortlist when we start sending.

Browse & filter

Skills finder

AI agent skills directory

Filter by kind or tag, search by name or repo, then open GitHub. Star counts refresh daily from GitHub; explainers stay curated. This is an editorial map — not a product catalog we sell.

10 skills on radar

#1Harness
ECC
affaan-m/ECC
★ 263kCross-harness operating system energy
What it doesAn OS-style playbook of reusable skills and safety hooks that keeps coding agents consistent across Claude Code, Codex, and Cursor.
Think of ECC as an operating system for coding agents: a big library of reusable skills, specialist agents, hooks, and safety checks that plug into Claude Code, Codex, Cursor, and friends. Instead of reinventing the same “plan → build → review → secure” loop every project, teams install a shared playbook. It is less a single prompt and more a harness that keeps agents consistent across tools.
- harness
- skills
- claude-code
- multi-agent
View on GitHub
#2Skill
Anthropic Agent Skills
anthropics/skills
★ 177kReference implementation of the skills format
What it doesAnthropic’s reference skills pack — SKILL.md folders that teach Claude how to load deep playbooks only when needed.
Anthropic’s public skills repo is the clearest “hello world” for the Agent Skills idea: folders with a SKILL.md, optional scripts, and progressive disclosure so the model only loads depth when needed. If you are writing your first skill or teaching a teammate what the format looks like, start here. It is also the closest thing to an official pattern book for how skills should be structured.
- skills
- SKILL.md
- claude
- spec
View on GitHub
#3Skill
Addy Osmani Agent Skills
addyosmani/agent-skills
★ 97.5kProduction engineering playbooks
What it doesProduction engineering checklists for coding agents — define, plan, build, verify, review, and ship with clear exit criteria.
A curated set of production-minded skills for coding agents — define, plan, build, verify, review, ship — written like checklists a senior eng would actually use. The value is specificity: clear exit criteria and verifiable steps instead of vague “be careful” advice. Great when you want agents to behave more like a disciplined teammate than a chatty autocomplete.
- engineering
- skills
- cursor
- codex
View on GitHub
#4Harness
Browser Use
browser-use/browser-use
★ 115kAgents that drive real browsers
What it doesLets AI agents click, type, and navigate real web pages for scraping, form-filling, and end-to-end testing.
Browser Use is the popular open stack for letting agents click, type, and navigate real web pages — the practical cousin of “computer use” demos. Teams use it to scrape, fill forms, and test flows without writing brittle Selenium scripts by hand. If your agent needs to leave the IDE and work on the live web, this is one of the default starting points.
- browser
- computer-use
- automation
- agents
View on GitHub
#5Skill
Cloudflare Security Audit Skill
cloudflare/security-audit-skill
★ 17.6kMulti-phase vuln hunting skill
What it doesTurns a coding agent into a structured security auditor with recon, hunting, validation, and machine-readable findings.
A coding-agent skill that turns your model into a structured security auditor: recon, parallel hunting, validation, reporting, and machine-readable findings. It is the single-repo version of ideas Cloudflare has written about for vulnerability harnesses. Useful when you want repeatable audits with evidence — not a one-off “please find bugs” prompt.
- security
- audit
- SKILL.md
- cloudflare
View on GitHub
#6Toolkit
Vercel Skills CLI
vercel-labs/skills
★ 32.1knpx skills add … across 40+ agents
What it doesThe package manager for Agent Skills — install and sync skill packs into Claude Code, Cursor, Codex, and more with one CLI.
The open CLI for installing and syncing Agent Skills across Claude Code, Codex, Cursor, OpenCode, and many more. One `npx skills add` command maps a GitHub skill pack into the right folders for each harness. If the skills ecosystem is the content, this repo is the package manager people keep recommending.
- cli
- skills.sh
- install
- ecosystem
View on GitHub
#7Toolkit
Vercel Agent Browser
vercel-labs/agent-browser
★ 42.9kBrowser tooling aimed at agents
What it doesGives coding agents a reliable browser surface so they can navigate and verify real UI behavior, not just source code.
Agent Browser is Vercel Labs’ take on giving coding agents a reliable browser surface for navigation and checks. It sits in the same family as browser-use, but oriented toward agent workflows and modern web app testing. Reach for it when you want agents to verify UI behavior instead of guessing from source alone.
- browser
- vercel
- agents
- testing
View on GitHub
#8Harness
CUA
trycua/cua
★ 25kComputer-use 2.0 infrastructure
What it doesInfrastructure for computer-use agents: open drivers, cross-OS fleets, and benchmarks beyond a single browser tab.
CUA focuses on scaling computer-use: open drivers, cross-OS fleets, and benchmarks for training and evaluation. Where browser-only tools stop at the tab, CUA aims at broader desktop and environment control. It is infrastructure for teams building or evaluating agents that operate real machines, not just chat APIs.
- computer-use
- drivers
- benchmarks
- fleets
View on GitHub
#9Harness
OpenAI Agents (Python)
openai/openai-agents-python
★ 29.6kOfficial Python agents SDK
What it doesOpenAI’s Python SDK for multi-step agents with tools, handoffs, and tracing — the programmable runtime side of the stack.
OpenAI’s Python Agents SDK is a harness for building multi-step agents with tools, handoffs, and tracing. It is the “framework” side of the stack: you write code that orchestrates models rather than dropping Markdown skills into a coding IDE. Pair it with skills when you want both procedural knowledge and a programmable runtime.
- openai
- sdk
- agents
- python
View on GitHub
#10Toolkit
json-render
vercel-labs/json-render
★ 17.1kGenerative UI from structured JSON
What it doesLets agents emit safe structured JSON that your app turns into real UI components instead of freeform HTML.
json-render is Vercel Labs’ generative UI framework: agents emit structured JSON, and your app turns that into real components safely. It is a skills-adjacent idea — teaching models to speak a constrained UI dialect instead of freeform HTML. Worth watching if your product is an agent that should draw interfaces, not just text.
- generative-ui
- json
- vercel
- ui
View on GitHub

FAQ

Skill vs harness, in plain English

Short answers so you can skim the map without jargon overload.

What is an agent skill?

A skill is a small, reusable playbook — usually a folder with a SKILL.md file — that teaches an AI coding agent how to do one job well (security audit, TDD workflow, docs update, and so on). The agent loads the skill when the task matches, instead of reinventing steps from scratch every time.

What is an agent harness?

A harness is the runtime and scaffolding around the model: tools, hooks, memory, permissions, multi-agent orchestration, and install paths. Skills are the recipes; the harness is the kitchen. Examples include ECC, OpenAI’s Agents SDK, and computer-use / browser stacks that let agents act outside the chat box.

How is this list picked?

v1 is a curated editorial shortlist of real GitHub projects in the agent-skills ecosystem. Use the finder to search by name, repo, or tags and filter by skill / harness / toolkit. We refresh star counts daily and write original plain-English “what it does” blurbs — we do not paste READMEs or claim ownership of any listed project.

Are these Intellure products?

No. Intellure is not affiliated with the listed projects. Radar is an editorial map so builders can discover skills and harnesses faster. If you want a managed AI front desk for WhatsApp and your website, that is a separate Intellure service.

Building for customers?

Need a managed AI on WhatsApp?

Radar maps open-source agent skills. Intellure is separate: a founder-led AI front desk that answers customers, captures leads, and books appointments on WhatsApp, Instagram DMs, and your website — set up once, run monthly.

Message on WhatsApp Contact

Get the daily top 5

Digest waitlist — plain-English picks of the hottest AI agent skills and harnesses. Not a newsletter blast dump; a short daily shortlist when we start sending.

Find AI agent skills — what each one actually does

Get the daily top 5

AI agent skills directory

ECC

Anthropic Agent Skills

Addy Osmani Agent Skills

Browser Use

Cloudflare Security Audit Skill

Vercel Skills CLI

Vercel Agent Browser

CUA

OpenAI Agents (Python)

json-render

Skill vs harness, in plain English

Need a managed AI on WhatsApp?

Find AI agent skills — what each one actually does

Get the daily top 5

AI agent skills directory

ECC

Anthropic Agent Skills

Addy Osmani Agent Skills

Browser Use

Cloudflare Security Audit Skill

Vercel Skills CLI

Vercel Agent Browser

CUA

OpenAI Agents (Python)

json-render

Skill vs harness, in plain English

Need a managed AI on WhatsApp?