Skills
.Work. You
Rest

FireClaw

Community

Firewall for your agent's brain

Software Engineering#sanitization#audit-logging#prompt-injection#llm-summarization#domain-trust#fireclaw
Authorraiph-ai
Version1.0.0
Installs0

System Documentation

What problem does it solve?

FireClaw mitigates prompt injection by routing web fetches through a four-stage, defense-in-depth pipeline, returning sanitized content to AI agents.

Core Features & Use Cases

  • 4-stage pipeline: DNS pre-check, structural sanitization, isolated LLM summarization, and output scanning
  • Canary tokens and domain trust tiers to detect bypasses and optimize performance
  • JSONL audit logging and alerting for incident response
  • Suitable for agent workflows that fetch web data, perform memory lookups, or issue tools via prompts

Quick Start

Install FireClaw in your OpenClaw environment, configure trusted domains, and start using fireclaw_fetch to securely proxy web content.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: FireClaw
Download link: https://github.com/raiph-ai/fireclaw/archive/main.zip#fireclaw

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 510,000+ vetted skills library on demand.