[01] SKILL
AI & LLMsOpenClaw
Indirect Prompt Injection
Detect and reject indirect prompt
by aviv4339
Install command
npx clawhub@latest install indirect-prompt-injectionOr install manually: copy to ~/.openclaw/skills/ (global) or <project>/skills/ (workspace)
Files in Indirect Prompt Injection/
How this skill works
When installed, this skill becomes available to your AI agents on Trooper. The agent can invoke it when relevant tasks are detected based on trigger phrases and context.
- Automatically activated when matching triggers are detected
- Runs within the agent's execution sandbox
- Follows workspace permission settings
- Can be combined with other installed skills
More AI & LLMs skills
4claw
4claw — a moderated imageboard for AI agents.
AI & LLMsView skill →
A2a
Decentralized Agent-to-Agent Autonomous Economy.
AI & LLMsView skill →
Aap Passport
Agent Attestation Protocol - The Reverse Turing Test.
AI & LLMsView skill →
Adaptive Suite
A continuously adaptive skill suite that empowers Clawdbot
AI & LLMsView skill →
Adversarial Prompting
Adversarial analysis to critique, fix.
AI & LLMsView skill →
Ag Model Usage
Use CodexBar CLI local cost usage to summarize
AI & LLMsView skill →