Trust center

Safety gates for reusable AI agent skills.

OpenAgentSkill does not treat every GitHub repository as equally safe. Each skill gets a safety gate before it can enter agent recommendations, install handoffs, and audit workflows.

4,432

Indexed

1,183

Auto allow

3,555

Reviewed+

Blocked

Verified

1,137

Strong audit, verified listing, install path, and no policy warnings.

View skills

Reviewed

2,418

Good audit signals. The agent can shortlist it, but may still require review.

View skills

Experimental

874

Useful for discovery, but not for autonomous installation.

View skills

Blocked

Excluded from default recommendations until a human reviews the risks.

View skills

Agent policy

Default to the safest useful skill, not the loudest repository.

The Resolve API ranks task fit, Trust Score, audit score, installability, permission hints, and policy limits together. Blocked candidates are returned only as audit context; they are not selected for agent installation.

Try Resolve API Browse audit reports

Policy contract

What an agent should do.

Verified

May install in a sandbox or low-risk workspace after normal review.

Reviewed

Can be shortlisted; require human approval if permission notes exist.

Experimental

Manual test only. Compare against safer alternatives first.

Blocked

Do not auto-install. Use only for explicit human audit workflows.

Safe shortlist

High-confidence skills to inspect first.

View verified

Codegraph

VERIFIED

Pre-indexed code knowledge graph for Claude Code, Codex, Gemini, Cursor, OpenCode, AntiGravity, Kiro, and Hermes Agent — fewer tokens, fewer tool calls, 100% local

94/100 safety99/100 trust44K stars

System Prompts And Models Of AI Tools

VERIFIED

FULL Augment Code, Claude Code, Cluely, CodeBuddy, Comet, Cursor, Devin AI, Junie, Kiro, Leap.new, Lovable, Manus, NotionAI, Orchids.app, Perplexity, Poke, Qoder, Replit, Same.dev, Trae, Traycer AI, VSCode Agent, Warp.dev, Windsurf, Xcode, Z.ai Code, Dia & v0. (And other Open Sourced) System Prompts, Internal Tools & AI Models

94/100 safety98/100 trust139K stars

Claude Mem

VERIFIED

Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More

94/100 safety98/100 trust82K stars

CrewAI

VERIFIED

Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.

94/100 safety98/100 trust53K stars

EasySpider

VERIFIED

A visual no-code/code-free web crawler/spider易采集：一个可视化浏览器自动化测试/数据采集/网页爬虫软件，可以无代码图形化的设计和执行爬虫任务。别名：ServiceWrapper面向Web应用的智能化服务封装系统。

94/100 safety98/100 trust44K stars

Gradio

VERIFIED

Build and share delightful machine learning apps, all in Python. 🌟 Star to support our work!

94/100 safety98/100 trust43K stars

CopilotKit

VERIFIED

The Frontend Stack for Agents & Generative UI. React, Angular, Mobile, Slack, and more. Makers of the AG-UI Protocol

94/100 safety98/100 trust35K stars

500 AI Agents Projects

VERIFIED

The 500 AI Agents Projects is a curated collection of AI agent use cases across various industries. It showcases practical applications and provides links to open-source projects for implementation, illustrating how AI agents are transforming sectors such as healthcare, finance, education, retail, and more.

94/100 safety98/100 trust32K stars

Nanoclaw

VERIFIED

A lightweight alternative to OpenClaw that runs in containers for security. Connects to WhatsApp, Telegram, Slack, Discord, Gmail and other messaging apps,, has memory, scheduled jobs, and runs directly on Anthropic's Agents SDK

94/100 safety98/100 trust30K stars

#10

Kestra

VERIFIED

Event Driven Orchestration & Scheduling Platform for Mission Critical Applications

94/100 safety98/100 trust27K stars

#11

NemoClaw

VERIFIED

Run agents like Hermes and OpenClaw more securely inside NVIDIA OpenShell with managed inference

94/100 safety98/100 trust21K stars

#12

Parlant

VERIFIED

Build reliable customer-facing AI agents with Parlant: an interaction control harness optimized for controlled, consistent, and predictable LLM interactions.

94/100 safety98/100 trust18K stars