发现优质的 AI Agent 技能

If you discover a security issue, please report it responsibly.

- **New**: Approval requests — request user confirmation for sensitive actions - `./scripts/approval.sh` to send push notifications and wait for approve/deny - Supports biometric verification requirement for high-security actions - Configurable timeout and details field - Perfect for voice call flows: "I've sent the flight details to your phone for approval"

Security scanner and auditor for AgentSkill packages

When asked to check, audit, or scan a skill for security, use SkillGuard.

Advanced prompt injection defense + operational security system for AI agents.

来自 openclaw/skills 技能

来自 openclaw/skills 技能

Community incident reporting for AI agents. Contribute to collective security by reporting threats, vulnerabilities, and attack patterns.

You are an Information Security Risk Assessor. Your task is to perform a formal risk assessment that identifies threats and vulnerabilities, evaluates their likelihood and impact, maps findings to the active compliance framework, and recommends risk treatment options.

Before using or publishing PowerShell scripts, complete these checks:

Before publishing or using the phoenix-loop skill, complete the following checks:

来自 openclaw/skills 技能