|
Best Claude Skills, MCP Servers, Plugins, AI Agents & Prompts Directory | ImAiFox
Discover and install 3,500+ Claude Skills, MCP Servers, Plugins, AI Agents, and Prompt templates. The largest open-source directory — verified from GitHub, u...
|
85 % |
|
ray.run: one endpoint for every MCP tool
Build MCP servers with your coding agent, run them beside remote MCP, OpenAPI, npm, and container services behind one endpoint, and publish them to your cust...
|
84 % |
|
skael — one registry for your team's AI skills, on every agent
Self-hostable platform for managing AI agent skills across engineering teams. Central registry, cross-agent sync, security scanning, and activation tracking.
|
84 % |
|
OpenClaw Skills Directory | Best ClawHub Plugins (ex-Moltbot)
Discover 2,800+ curated OpenClaw (formerly Moltbot) skills. Browse verified AI agent tools, integrations like Obsidian and Outlook, and safe plugins from the...
|
84 % |
|
AgentOpsSec — Secure your agents before they touch your tools
AgentOpsSec is the open-source security and observability stack for AI agents. Local-first, MIT-licensed CLI tools that audit MCP servers, control tool calls...
|
84 % |
|
Agent Plugins: The Portable Agent Plugin Standard
A developer field guide to Agent Plugins 1.0: package structure, skills, MCP servers, compatibility, security boundaries, and a working starter.
|
84 % |
|
Foo Guard — AI Agent Security Scanner
Scan agent configs and skills, share reviewed configuration versions in Library, enforce CI policy, and control supported requests with Hosted Gateway.
|
83 % |
|
Nox — The security scanner that understands your AI app
Open-source security scanner with first-class AI app security: prompt injection at the call site, embedding leakage, agent over-privilege, MCP hardening, cro...
|
83 % |
|
OpenAgere: terminal-native AI coding agent
Terminal-native AI coding agent for safe, extensible software engineering automation.
|
83 % |
|
PluginSec — AIDR for the plugin supply chain
The lightweight agent that inventories every IDE extension, browser extension, and AI-agent MCP server across your fleet — and stops the malicious ones befor...
|
83 % |
|
Daniel Kiska - senior full-stack · security · data/AI engineer
Senior full-stack, security and data/AI engineer in London. Interactive CV, verified case studies and technical writing.
|
82 % |
|
BacklogHQ | BacklogHQ
Open-source tools for AI agents. MCP-first, event-driven, crash-safe.
|
82 % |
|
DotAgent / Rules and skills for coding agents
Rules and skills for coding agents. One rules file and 21 skills for Claude Code, Codex, Cursor and other agents, installed through AgentSync.
|
82 % |
|
Free security Scanner + continuous security Agent
Scan files and URLs for free, then connect CypherScan Agent to continuously monitor websites, GitHub, WordPress, S3, CI/CD, and API or plugin workflows.
|
82 % |
|
ClaudePluginHub - The Largest Claude Code Plugin Directory
Get ranked Claude Code plugin recommendations for your stack, or search thousands of plugins, commands, agents, skills, and MCP servers before you install.
|
82 % |
|
VibeTesting Agent — security scans for vibe-coded apps
Black-box security for AI-built software. Free lite scan on the homepage. Hosted plans: Casual Coding $25 · Business Prototyping $100 · Mission Critical $250...
|
82 % |
|
AgenticStore — Free MCP Tools for AI Agents
31 free, open-source MCP tools for Claude, Cursor & Windsurf. Secure prompts, scan secrets, search the web — self-hosted, no account needed.
|
82 % |
|
Nullcone — Security that installs in one command
Free threat intelligence for developers and AI agents. Blocks malicious IPs, domains, prompt injections, and AI skill attacks. 1.56M+ threats. pip install or...
|
82 % |
|
Longxia — Static security scans for AI agent skills
Inspect public GitHub agent skills for risky instructions, scripts, permissions, and supply-chain behavior before installation.
|
82 % |
|
RepoGuard — Scan before you install
Scan any skill, plugin, or package for malicious code before you install it. Static security analysis with clear RED/YELLOW/GREEN verdicts.
|
82 % |
Detected social media networks