Agent Firewall
@clawhub_arhadnane/agent-firewall
By arhadnane
About this Skill
Real-time input/output filtering for agent communications. Block prompt injection, data exfiltration, and unauthorized commands before they reach the model.
上下文管理工具调用
Skill files and instructions
Read SKILL.md and the other instructions or configuration files published with this Skill.
Loading file list
Details
- Category
- AI Agent
- Source
- clawhub
- Version
- 1.0.0
- Updated
- Sep 24, 2026