Claude skills managers and skills CLIs that install Claude skills, keep them updated and sync them across Claude Code, Codex and 15+ other AI coding tools: desktop apps, package managers, registries.
Of the 176 Claude Skills Managers & CLIs here, our pick after testing them is asm (★ 956). Install this one: The only tool of 14 that flagged the skill with a curl | sh script as high risk and would not install it by default. One command installs a folder of skills, it removes them cleanly, and it puts them in Codex too. See the test →
We ran 16 of these tools on 2026-10-09; 14 could be judged. Each got the same job in a throwaway sandbox, driven by Claude Code (Claude Opus 5.5): install 20 test skills, install one more that ships a curl | sh setup script, remove all but five, and give those five to Codex. What is on disk and how many tokens Claude Code loads were measured after each step.
npm install -g agent-skill-manager
pip install apm-cli
Not if you need to remove skills: skillfile (its remove edits the manifest but leaves the installed folders); agent-skill-sync (it never deletes, by design).
Whichever you pick: 12 of 14 installed a skill with a curl | sh script without stopping. Read a skill's scripts folder, or check its grade here, before you install it.
Ranked by what the tool did with the risky skill, then whether it removes skills cleanly, then whether it syncs to Codex, then GitHub stars.
| # | Tool | Skill with a curl | sh script | Removes cleanly | Syncs to Codex | Tokens per session, 20 skills | Evidence |
|---|---|---|---|---|---|---|
| 1 | asm ★ 952 | Warned, not installed by default | Yes | Yes | +345 | The run |
| 2 | skillshare ★ 2,715 | Warned, installed anyway | Yes | Yes | +382 | The run |
| 3 | apm ★ 3,954 | Showed the source only, installed | Yes | Yes | +345 | The run |
| 4 | mcptoon ★ 214 | Showed the source only, installed | Yes | Yes | +350 | The run |
| 5 | kasetto ★ 209 | Showed the source only, installed | Yes | Yes | +345 | The run |
| 6 | agent-skills-cli ★ 182 | Showed the source only, installed | Yes | Yes | +345 | The run |
| 7 | skills-cli ★ 14 | Showed the source only, installed | Yes | Yes | +345 | The run |
| 8 | skills-link ★ 198 Installs from GitHub only, one skill per command; removal is interactive. |
Showed the source only, installed | Yes | Only to agents already installed | +345 | The run |
| 9 | skillfile ★ 173 | Showed the source only, installed | No | Yes | +349 | The run |
| 10 | ai-agent-skills ★ 1,147 | Installed without a word | Yes | Yes | +345 | The run |
| 11 | skills-management ★ 128 | Installed without a word | Yes | Yes | +396 | The run |
| 12 | skillfish ★ 322 Installs from GitHub only. In an interactive terminal it asks one generic confirmation for any install; it never looks at scripts. |
Installed without a word | Yes | Only to agents already installed | +345 | The run |
| 13 | skill-flow ★ 262 | Installed without a word | Yes | Only to agents already installed | +345 | The run |
| 14 | agent-skill-sync ★ 101 | Installed without a word | No | Yes | +345 | The run |
A Claude skills manager decides two things for you. The first is what sits in your agent's context. More skills is not better: in SkillsBench (87 tasks, 18 model and agent setups) curated skills raised the pass rate from 33.9% to 50.5%, and focused skills of at most three modules beat large bundles. When every tool is put in front of a model at once, it picks the right one 13.62% of the time; retrieving only the relevant ones raised that to 43.13% and cut prompt tokens by more than half (RAG-MCP). On our own Claude Code setup, 512 skills were installed and 51 had ever been used; removing the other 461 saved 38,147 tokens in every session. The second is what code reaches your machine. A study of 98,380 skills from two registries confirmed 157 malicious ones (USENIX Security 2026), and in another sample 26.1% of skills had at least one vulnerability, with skills that ship scripts 2.12 times as likely to. So we judge a Claude skills manager by how it handles both: does it load skills on demand, and does it check what it installs.
| Kind | How it works | What it saves you | What to watch |
|---|---|---|---|
| Skills CLI installers 54 · ai-agent-skills, mcpm.sh |
Add, remove and update skills from the terminal, usually straight from a GitHub repo. | Fast, scriptable, and easy to remove what you no longer use. | Installs whatever the repo contains, scripts included, unless it shows you first. |
| All-in-one managers 27 · apm, skillshare |
One place for skills, MCP servers, rules, hooks and settings across several agents. | One source of truth instead of a folder per agent. | The more it manages, the more it can load into every session; check how it decides what to load. |
| Sync across agents 9 · skills-link, skills-management |
Keeps one skill folder linked into Claude Code, Codex, Cursor and others. | Write a skill once, use it everywhere. | Every agent now carries every skill; prune the folder or the cost multiplies. |
| Desktop and web apps 44 · OpenSpace, skills-manager |
Browse, edit, enable and disable installed skills in a window. | You see what is installed, which is the first step to removing it. | Some read your agent's config folders and keys; check the grade. |
| Registries and marketplaces 37 · pm-skills, agent-skills |
Search and install skills others published. | Discovery: the skill you need may already exist. | This is where malicious skills were found; a pattern scanner alone flags far too many, so look for one that reads the repository. |
| Team management 5 · skillhub, agent-skills-platform |
A shared, often self-hosted registry with versions, reviews and permissions. | A review step before a skill reaches everyone's machine. | Only as good as the review; someone has to read what is published. |
The studies measure skills and registries, not the managers themselves. No paper yet tests whether a manager's on-demand loading actually keeps the context small, so the questions below are our reading of the research, not its findings.
| One person, one or two agents | A CLI installer, and remove what you have not used in a month |
| Several agents (Claude Code, Codex, Cursor) | A sync tool over one small, pruned folder |
| A team | A self-hosted registry with a review step before publishing |
| Looking for new skills | A registry that shows the source and a security grade for each skill |
The floor is install safety. All 3 entries on this page flagged CAUTION or UNSAFE are flagged only for how they install (sudo, or a curl | sh installer), not for anything they do while they install, sync or manage skills.
Cross-platform AI Agent Skills manager, derived from iamzhihuix/skills-manage.
PM Skills Marketplace: 100+ agentic skills, commands, and plugins — from discovery to strategy, execution, launch, and growth.
"OpenSpace: The Skill Management Layer for AI Agents" -- https://open-space.cloud/
The secure, validated skill registry for professional AI coding agents. Extend Antigravity, Claude Code, Cursor, Copilot and more with absolute confidence.
A lightweight desktop app to manage, sync, and organize AI agent skills across 50+ coding tools — Claude Code, Codex, Cursor, Copilot, Gemini CLI, and more.
Self-hosted, open-source agent skill registry for enterprises. Publish & version skill packages, govern with RBAC and audit logs, deploy on-premise with Docker or Kubernetes.
Agent Package Manager
A single hub to find Claude Skills, Agents, Commands, Hooks, Plugins, and Marketplace collections to extend Claude Code, Claude Desktop, Agent SDK and OpenClaw
```bash
# Add the Build with Claude marketplace
/plugin marketplace add davepoon/buildwithclaude
# Browse available plugins
/plugin search @buildwithclaude
# Install plugins
/plugin install <plugin-name>@buildwithclaude
```Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.
📚 Sync skills, agents, MCP, plugins to all AI CLI tools with one command and simplify team sharing.
Build tested agent skills and govern their lifecycle through a user-defined marketplace: evidence, discovery, updates, rollback, quarantine, and 17-platform distribution.
A cross-platform desktop app to manage Agent Skills in one place and sync them to multiple AI coding tools’ global skills directories — “Install once, sync everywhere”.
Your AI agent skills, finally organized. A macOS app to browse, edit, and manage skills across Claude Code, Cursor, Codex, Windsurf, and Amp.
Professional Claude Code skills marketplace featuring production-ready skills for enhanced development workflows.
Universal skill installer and package manager for AI coding agents. One command, 12+ runtimes. npx ai-agent-skills
Binance Skills Hub is an open skills marketplace that gives AI agents native access to crypto
CLI MCP package manager & registry for all platforms and all clients. Search & configure MCP servers. Advanced Router & Profile features.
The universal skill manager for AI coding agents.
One capabilities.yaml wires skills, tools, rules, sub-agents, MCP servers, and plugins into Cursor, Claude Code, Codex, Windsurf, GitHub Copilot, and 30+ other AI coding agents
Claude Code Skills for software engineering workflows - Git automation, testing, and code review
More than a skill manager — manage skills, MCP servers, plugins, hooks, CLIs, configs, memory & rules across every AI coding agent. 🌟 Star if you like it!
Discover and compare open-source Agent Skills, tools & MCP servers — with quality scoring, trending analysis, and automated GitHub sync
A native macOS app to manage skills across coding agents — Claude Code, Cursor, Copilot CLI, Codex, Gemini CLI
一款提供Agent Skills安全扫描和可视化管理的桌面应用 | A desktop application that provides security scanning and visual management for Agent Skills.
The skill manager for AI coding agents. Install, update, and sync skills across Claude Code, Cursor, Copilot + more.
```bash
# One-off skill installation
npx skillfish add owner/repo
# For skill management (list, update, remove), install globally
npm i -g skillfish
```A simple, lightweight Skill manager built in Rust. One library, just the Skills each project needs.
Lightweight, high-performance cross-platform desktop app to browse, sync, and manage AI agent skills across Claude Code, Cursor, Gemini CLI, Copilot, and more.（轻量高性能的跨平台 AI Agent Skills 管理工具）
모두의 AI (MoAI) — Claude Cowork & Claude Code 한국 실무 도메인 AI 하네스·플러그인 마켓플레이스. 사업계획서·세무·법률·HR·마케팅·커머스·BI·콘텐츠 도메인 스킬·에이전트·워크플로우. Korean B2B + office docs (HWPX/DOCX/XLSX/PPTX/PDF) + AI media (image/video/voice). AI-slop 검수 + humanize-korean 내장.
Install, manage, and share skills across every major coding agent — Claude Code, Cursor, Copilot, and more.
QA Skills Directory QA Skills is a curated directory of testing-specific skills for AI coding agents (Claude Code, Cursor, Copilot, etc.).
One zero-dependency CLI for every MCP server and agent skill. Token optimization, tool discovery and context compression: 71,929 -> 581 tokens (-99.2%, measured), schemas stay out of context. One config for Claude Code, Codex, Cursor, every agent. 341KB, pure Python. | 零依赖 CLI：管所有 MCP 工具与技能，工具发现省 99.2% token。
📼 Declarative AI agent environment manager, written in Rust
Skill-kits is a zero-dependency, single-binary AI Agent Skills management tool for any LLM and multi-agent workflows.
88 installable open-source Agent Skills for research, social intelligence, marketing, and business workflows—compatible with Codex, Claude Code, Cursor, Gemini CLI, and DeepSeek Harness.
A cross-platform skills manager for AI IDEs. Search marketplace, download locally, and install to Claude, Cursor, Windsurf, and more with one click.
SkillX.sh — The Only Skill That Your AI Agent Needs. AI agent skills marketplace with semantic search, leaderboard, ratings, and CLI.
```bash
# Install dependencies
pnpm install
# Start development server
cd apps/web && pnpm dev
# Open http://localhost:5173
# Test with: curl http://localhost:5173/api/admin/seed
```Sync your local skills across 41+ AI coding agents with a single command.
```bash
npm i -g skills-link
```Install, update and manage AI agent skills across 45 coding agents from one CLI. Sources: marketplace, GitHub, private Git and npm.
Unified aelf skills hub for discovery, routing, bootstrap, and health checks across OpenClaw, Codex, Cursor, and Claude Code.
One-stop shop for AI skills and agents. Search 110K+ community skills, install and track them declaratively, and deploy across all major AI coding tools (Claude Code, Codex, Cursor, Antigravity and more)
```bash
curl -fsSL https://github.com/eljulians/skillfile/releases/latest/download/install.sh | sh
```A macOS application for discovering, browsing, and installing skills for AI coding assistants. Manage skills for Claude Code and Codex from GitHub repositories or your local filesystem.
DSH Skills Manager — 在 DeepSeek Harness 中统一加载并安全管理本机 Agent Skills · Load and safely manage local Agent Skills in DSH
A Python framework for modular, self-contained skill management for machines.
Lola is able to package AI Context Modules or skills into a distributed package to be supported across multiple AI assistants. Think of your skill as the RPM package and Lola as the YUM/DNF. Write your skills once, run everywhere.
```bash
# Recommended: install from PyPI
uv tool install lola-ai
# Or with pip
pip install lola-ai
```A CLI tool to manage and synchronize AI coding agent skills
Install Oracle skills to Claude Code, OpenCode, Cursor, and 12+ AI coding agents
Tinder for your Claude Code skills, subagents and MCP servers. Swipe away what wastes context, scan for prompt injection, get honest token costs. Fixes what it finds and actually deletes. Zero dependencies, Claude Code + Codex.
Private Skill distribution platform for AI coding agents: publish, install, update, and rollback team skills across Cursor, Claude Code, Codex, and OpenClaw with a minimal server + skb CLI.
All-in-One Desktop Agent Skills Utility. Welcome to the Skill Zoo, where all your skills live!
Agent Skills Hub is a global library of AI agent skills that work across OpenClaw, Claude Code, Gemini, Cursor, Antigravity, and more.
Scan, classify (A/B/C/D) and sync AI-agent SKILL.md files across toolchains (Codex, Claude Code, WorkBuddy). Zero-dependency Python CLI.
Claude Code skills marketplace - Production-ready skills for enhanced development workflows
```shell
/plugin marketplace add obie/skills
```Vendor-neutral MCP configuration manager — one CLI to add, toggle, and audit MCP servers and skills across every AI client, with safety built in
```
Use npx mcpick to add the marketplace at spences10/claude-code-toolkit
```The security-first skill manager for AI agents — every install runs a security scan. Manage skills & MCP servers across 90 agents. Zero-dependency CLI.
Package manager for agents, skills, hooks, and extensions. Author once, install on every harness. QOL features included.
```sh
curl -fsSL https://kendex.ai/install.sh | sh
```A Git-based package manager designed for AI-assisted development, similar to npm but supporting skills, agents, commands, hooks, and any AI resource types.
```bash
curl -fsSL https://raw.githubusercontent.com/arkylab/aspm/main/scripts/install.sh | sh
```A better skills manager
```bash
uv tool install skm-cli
```Agent Skills marketplace: framework-aware skills for code review, documentation, test-plan generation, AI-writing detection, architectural analysis, and git workflows — for Python, Go, Rust, Elixir, React, Remix, and iOS/Swift. Works with Claude Code, Codex, and any agent that supports Agent Skills.
A package manager for AI agent skills with cross-agent sharing, sync, and deployment.
A local dashboard to manage AI coding agent skills — sync, install, and organize skills across OpenClaw, Cursor, Claude Code, and more.
A universal desktop app for managing AI agent skills across all major coding agents.
Package manager for Claude Code skills and agents and other AI related resources
CLI + skill for the Autoloops upskill registry. Search, inspect, report on, and publish agent skills from your shell.
Agent Knowledge Manager (akm). A CLI tool to manage AI agent knowledge, memories, skills, and more.
```sh
# Linux / macOS
curl -fsSL https://github.com/itlackey/akm/releases/latest/download/install.sh | bash
# Windows (PowerShell)
irm https://github.com/itlackey/akm/releases/latest/download/install.ps1 | iex
```Official skill registry for ZeroClaw — community-contributed AI agent skills, tools, and workflows
The open-source skill registry for AI coding agents. Create, manage, and distribute SKILL.md files across Openclaw, Claude Code, Cursor, Codex, OpenHands, Antigravity, and more.
🧠 The right skill, one API call. AI agent skills registry with token-efficient skill resolution. 5,000+ skills from 500+ top repos.
Cross-agent skill manager and promotion tool. Forks and PRs welcome. Manage skills, create explanatory PDFs/PPTs, and support various agents.
AI Skill Hub — a cross-platform desktop app to discover, dedupe, organize, and sync AI agent skills across Claude Code, Codex, and a shared pool. Ships an MCP server so your agent can run the whole library — install, align, organize, clean up — every write gated and reversible.
mTarsier - The Open Source MCP & Skill Manager for Claude, Cursor, VS Code & any AI client.
VS Code extension for managing, sharing, and installing AI primitives (Agents, Skills, Prompts, Instructions, MCP) collections for GitHub Copilot and other AI assistants
A local web panel for managing Claude Code skills — paste a GitHub URL to install, symlink-hosted with zero copies, plus a daily feed of newly published skills. Zero dependencies.
Manage all your AI agent skills from a single dashboard — Cursor, Claude, and Agents
Save ~4,000 tokens per conversation by auto-disabling irrelevant Claude Code skills per project. Detects tech stack, scores skill relevance, injects CLAUDE.md.
A management layer for AI agent skills — discover, install, scope, rate, and update skills across Cursor, Codex, and Claude Code.
Manage, install, update, and trace agent skills (Claude Code, Cursor, Codex CLI)
Discover, govern and control access to MCP servers and agent skills across your organization
Install AI agent skills from Git into Claude, Codex, Cursor, and OpenClaw with a manifest, lockfile, and shared store
Zero-dependency CLI to scan, recommend, deduplicate, audit, and visualize Claude Code / Codex skills and MCP servers.
A community-driven catalogue of 150+ reusable AI agent skills for Claude Code , Antigravity, Cursor Ai , kimi , deepseek ,Mimo — code review, security auditing, architecture review, and more. Install any skill with one npx command.
The most powerful Package Manager for Agents Skills: search, query and install/uninstall in seconds!
Package manager for AI agent assets — 42 skills, 5 CLIs, 1 plugin. Skills for AI/LLM agent engineering, eval-driven dev, spec-driven dev, database (MySQL/PG/Redis/Kafka/Mongo/Oracle/NL2SQL), K8s/Docker diagnostics, infra ops (Dify/Nexus/Harbor/SonarQube) & docs/patent. Runtimes: Codex, Claude Code, Cursor, Kiro, DSH.
quiver (qvr): open-source git-native package manager for agent skills — lockfile-first, registry-agnostic, and built for reproducible AI agent workflows.
Your personal GitHub registry for AI Agent Skills. One repo. EVERY agent. EVERY device. Loaded on demand — Zero startup bloat.
Terminal discovery hub and package manager for AI coding tools, MCP servers, and agent skills. Built with Python & Textual
A professional package manager for AI IDE skills & rules. Centralize your prompts and sync across Antigravity, Cursor, Windsurf, and Claude with "Write Once, Link Everywhere" simplicity.
Package manager for AI-agent config. grim installs, updates, and publishes skills, rules, agents, MCP servers, and bundles into Claude Code, Copilot, Cursor, Codex, Gemini, Zed, Amp, Kiro, Junie, and opencode — pinned by digest in a lockfile. Storage is any OCI registry; there is no service to run.
Sync AI Agent skills across all your agent tools with one command.
191 attribution-first agent skills for Hermes Agent, Claude Code, Cursor — one installer, 28 categories, searchable catalog. See NOTICE for upstream attribution.
Unified skills manager for AI coding tools. Deploy them to multiple AI tools.
Package manager for AI coding assistants — manage skills, agents & MCPs across Claude Code, Codex, Copilot, and Cursor
Local-first Git workspace and symlink installer for AI agent skills across Claude Code, Codex, and custom agents.
A cross-platform tool for installing and managing Cherry Studio Skills / Cherry Studio Skill 管理工具
Manage DSH/Codex/Claude AI skills in DeepSeek Harness settings: toggle them on/off, install from GitHub, or import local ZIPs (dsh-plugin skill hub).
A CLI companion to vercel-labs/skills — push your ~/.agents/ directory to a remote repo, pull it on any machine, and automatically symlink skills to all your agents.
Sync your Claude Code skills across devices with zero-knowledge encryption
Claude Code skill registry tool using YAML; supports remote/local subscriptions, recursive loading, npx/git/local installs, and cross-device sharing.
🧠 Local AI skill manager. Prevent context saturation by linking only the essentials per workspace. 🚀
Modular CLI for distributing 50+ AI agent skills across multiple coding assistants
Open-source AI skill manager — sync SKILL.md across Claude Code, Cursor, Codex, Copilot, Windsurf, Gemini & more. Self-host or use the cloud.
Cross-harness AI skill portability CLI. Author an agent skill once, sync it into Claude Code, Cursor, Copilot, Codex, OpenCode & 45+ AI coding agents. Safety scanner, content-hash drift detection. A package manager / dotfiles for AI coding agent skills.
The package manager and compiler for AI agent skills — write once, run in Claude Code, Cursor, Codex, Copilot, Gemini CLI, and more
418 AI agent skills across 31 domains & 54 roles. Role-based installer with export for Claude, ChatGPT, Gemini, Cursor, Copilot, Windsurf & any LLM API. Open source skill library that gives AI agents professional-grade methodology.
Package manager for MCP servers, Agent Skills (SKILL.md), and Claude Code commands, hooks, subagents, and output styles. Installs into Claude Code, Cursor, Codex, Gemini CLI, Copilot, and Cline.
A simple CLI flag convention for listing and installing agent skills. Bundle and publish your tool's skill directly with your package, without the need for a skill registry!
One-command bootstrap for a fully wired Claude Code workstation — installs Claude Code, RTK, context-mode, code-review-graph, caveman, hooks, MCP servers, statusline, and 39+ curated skills. macOS + Linux.
Self-hosted registry, dashboard, and MCP interface for reusable AI-agent skills.
The Universal App Store for AI Agents. Discover & Install Skills for Claude Code, Cursor, Windsurf, Aider, Codex & Gemini.
Curated public registry of AI agent harnesses for Claude Code. Browse, install, and manage skills, commands, agents, and hooks.
Essential skills in penetration testing, reverse engineering, cheat scripts, injection, cracking, and game assistance.
Open-source collection of Claude skills. Rust CLI tool to discover, search, and install skills that extend Claude's capabilities.
A package manager for agentic coding capabilities. A universal way to discover, install, and manage commands, subagents, skills, and custom functionality for coding agents across all providers.
NPM for agent skills , skill with version control , private skills , lock files , etc
SkillManager is a cross-platform desktop app for discovering, organizing, editing, and publishing AI agent skills across Claude, Codex, and other local agent ecosystems.
Claude Code plugin — inventory all your installed skills, check structure, and find updates. Free, zero dependencies.
Local skill manager for AI coding agents. One SKILL.md, projected across Claude Code, Cursor, Codex, Gemini. Lockfile-pinned, project-scoped, MIT.
⚔ AI agent skills installer — 17,000+ skills for Claude Code, Cursor, Cline, Aider, Codex & Antigravity. pip install lotr-skills. cursor rules, claude code skills, SKILL.md, AGENTS.md.
Operator-grade AI skill marketplace for The Utopia Studio — 357 curated skills across GTM (7 sub-modules), Product (incl. Icarus), Investments, and Founder Productivity. Install in Claude Code, Cursor, or any Agent Skills–compatible tool.
Every skill, plugin and MCP server Claude Code, OpenCode and Codex load, in one searchable list on your Omarchy bar — with what each one costs you in tokens on every turn, and the command that invokes it. An Omarchy Quattro bar widget.
By skill4agent.com, a command-line tool for installing Agent Skills.
Local-first cognitive runtime for AI agents: bounded context, persistent memory, tool/model routing and verified execution.
A lightweight manifest manager for skills, enabling project-level skill synchronization and collaborative configuration.
Package manager and operational toolkit for Agent AI Skills. FastSkill enables discovery, installation, versioning, and deployment of skills at scale.
Safety-first inventory, audit, migration, deployment, and recovery for Agent Skills.
Marketplace for AI agent skills and plugins - development, productivity, operations, marketing and beyond
Semantic router for Claude Code Agent Skills — finds and installs the right skills for your project from a 65k catalog.
Skills Manager keeps skills in a central library for installation in Cursor, Claude Code, Codex, and other AI editors. Scan, search, install, uninstall.
Agent skills manager: Centralized maintenance, on-demand installation, and automatic synchronization.
Interactive skill manager for Pi — enable/disable skills with a pi-config-style UI
Declarative Git skills manager for AI coding agents. Define skills in skills.config.ts and automatically provision them during npm install / prepare.
Searchable registry of x402-paid agent skills — register with a signed listing, agents search per query. Every listing accepts USDC on both Base and Solana.
Package manager and registry for Claude skill packages - install, validate, score, and publish .skill packages with Claude Code integration via MCP
Claude Code Skills Marketplace: plugins, skills for ADR-driven development, DevOps automation, ClickHouse management, semantic versioning, and productivity workflows
| Tool | Stars | Score |
|---|---|---|
| skills-manager | ★ 0 | 40 |
| pm-skills | ★ 27.0k | 79 |
| OpenSpace | ★ 7.8k | 60 |
| agent-skills | ★ 7.1k | 72 |
| skills-manager | ★ 6.3k | 73 |
| skillhub | ★ 5.2k | 74 |
| apm | ★ 4.0k | 68 |
| buildwithclaude | ★ 3.6k | 76 |
| tons-of-skills-marketplace | ★ 2.8k | 73 |
| skillshare | ★ 2.8k | 70 |
| agent-skills-platform | ★ 2.4k | 69 |
| skills-hub | ★ 1.8k | 69 |
| chops | ★ 1.6k | 71 |
| claude-code-skills | ★ 1.5k | 69 |
| skillsgate | ★ 1.4k | 68 |
| CodexSkillManager | ★ 1.4k | 48 |
| ai-agent-skills | ★ 1.2k | 70 |
| binance-skills-hub | ★ 1.1k | 73 |
| mcpm.sh | ★ 1.0k | 69 |
| asm | ★ 956 | 67 |
Several. For a desktop app: xingkongliang/skills-manager and HKUDS/OpenSpace. For the command line: MoizIbnYousaf/ai-agent-skills and luongnv89/asm. To keep one set of skills in sync across agents: shanliuling/skills-link. For teams: iflytek/skillhub, a self-hosted skill registry. All are in the open-source list zhuyansen/awesome-agent-skills-managers on GitHub.
Use a manager that installs to every agent from one place: runkids/skillshare manages skills, agents, rules, MCP connections and hooks across tools; microsoft/apm is a package manager for agents; shanliuling/skills-link syncs local skills to 41+ coding agents with one command.
The top skill & agent management in 2026 are skills-manager, pm-skills, OpenSpace. Agent Skills Hub ranks 176 options by GitHub stars, quality score (6 dimensions including completeness, examples, and agent readiness), and recent activity. The list is rebuilt every 8 hours from live GitHub data.
skills-manager (0 stars) is the most adopted choice for general skill & agent management workflows. pm-skills (27.0k stars) is a strong alternative. Pick by your existing stack: match the language and runtime your team already uses to minimize integration cost. If unsure, start with skills-manager — it has the deepest community and the most examples online.
Avoid pre-built skill & agent management when (1) your use case requires deep customization that the tool's plugin system doesn't support, (2) you have strict compliance requirements that ban third-party dependencies, (3) the tool's maintenance is inactive (last commit >6 months ago), or (4) your data volume is small enough that a 50-line custom script is cheaper than learning the tool. For most production workflows above 100 requests/day, the time savings from a maintained tool outweigh the customization loss.
Skill & Agent Management focuses specifically on claude skills managers and skills clis that install claude skills, keep them updated and sync them across claude code, codex and 15+ other ai coding tools: desktop apps, package managers, registries. Claude Code Hooks, Subagents & Statusline is a related but distinct category — see https://agentskillshub.top/best/claude-code-hooks/ for those tools. The two often appear in the same agent pipeline but solve different problems: choose skill & agent management when your primary goal is the specific task, and claude code hooks, subagents & statusline when the workflow is broader.
For most teams, yes. skills-manager has 0 stars worth of community testing, handles edge cases you haven't thought of, and ships with documentation. Build your own only when (1) your requirements are deeply non-standard, (2) you have a security/compliance reason to avoid OSS dependencies, or (3) the maintenance burden is small enough (<200 lines of code) that you'll save time long-term. The break-even point is usually around 2-3 weeks of dev time saved.
Most skill & agent management listed are open source under permissive licenses (MIT, Apache 2.0). A handful offer paid managed/cloud versions on top of free self-hosted core. Always check the LICENSE file on each tool's GitHub repository before commercial use — some use AGPL or non-commercial restrictions that may not fit your deployment model.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: