Prompt Injection Security

What you can ask

• Review this system prompt and retrieved-content workflow for injection risks.Copy

  · Injection checker skill

• Threat-model prompt injection in this AI agent and its tools.
• Create safe regression tests for these prompt-injection concerns.

Skills

4 in the ChatGPT listing

• Agent tool boundariesReview AI agent tools, permissions, arguments, outputs, and action approvals for prompt-injection-driven misuse paths.
• Indirect injection reviewExamine documents, web pages, retrieval chunks, messages, and multimodal inputs as possible indirect prompt-injection sources.
• Injection checkerOrchestrate evidence-based prompt-injection risk reviews across prompts, retrieval, multimodal input, memory, and agent tools.
• Mitigation regression testsTurn prompt-injection findings into defense-in-depth fixes and safe repeatable regression tests.