Agent Skills Evaluation Framework

Help your agents create better skills

An evaluation and evolution tool for Agent Skills.

[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.

Evaluate agent skill quality. Find the weakest link. Fix it. Prove it worked.

Multi-tier framework for evaluating AI agent skills with quality gates, semantic overlap detection, synthetic evaluation dataset generation, and live agent evaluation that measures how skills affect agent behavior.

A test runner for agentskills.io-style AI agent skills

AAS Core is the local, agent-first control plane for complete catalog discovery, agent-owned selection, stack validation, and planning, backed by 2,115+ agentic skills. Includes CLI, local MCP, catalog, plugins, and Workbench.

QA Skills Directory QA Skills is a curated directory of testing-specific skills for AI coding agents (Claude Code, Cursor, Copilot, etc.).

CLI that learns repository-specific Claude skills with evolutionary search.

SRA-Bench and SR-Agents: a benchmark and toolkit for skill-retrieval-augmented LLM agents.

Use cultivar to test your Agent Skills, run them in sandboxes, and across different agents.

🪨 why use many token when few token do trick — Claude Code skill that cuts 65% of tokens by talking like caveman

Professional-grade AI coding toolkit: 94 skills, 44 agents, multi-platform (Claude, Cursor, Windsurf, Copilot, Gemini, Cline, Roo Code, Aider, Augment, Antigravity, Codex CLI, opencode).

🤖🤮 Hate the yuck codes that agents generated? Try `alint`, a ESLint like toolchain for intent driven code check, freeze your skills, AGENTS.md to lint rules

TeamCity from your terminal – or your AI's. Builds, logs, agents, agent terminals, queues.

Tiny Rust port of Pi — a ~4 MB statically-linked coding agent CLI for old/low-resource Linux boxes. OpenAI + Anthropic providers, streaming tool calls (read/write/edit/bash), JSONL sessions, ratatui TUI, Agent Skills, Claude Code-style hooks, -p print mode.

Interactive terminals for AI agents, built for what you can't --yes away. SSH+MFA, GRUB/U-Boot, debconf installers, SOL/serial consoles, fsck, cryptsetup, pdb/gdb, apt, certbot, pwsh and even Vim in tmux-backed sessions. Agent-driven, human-assisted for secrets/MFA. Single-file Python. Agent Skill. CI with 700+ tests. BSD License.

Companion products for AI-assisted software work.

Anti-detect browser automation CLI & Skills for AI agents — Camoufox-powered fingerprint spoofing, no bot-detectable Playwright leaks

A framework-agnostic, git-native standard for defining AI agents

Delegate tasks to DeepSeek right inside your Claude Code / Codex sessions.

Authorized security testing workspace. v2 TypeScript terminal product on release; v1 Python on pypi-release; Go branch is demo only.

The control plane for AI coding agents.