knowledge-extractor
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- PROMPT_INJECTION (LOW): The skill is susceptible to indirect prompt injection. Ingestion points: Conversational history, debugging logs, and problem-solving context (README.md). Boundary markers: Absent; there are no instructions provided to the agent to distinguish between data and embedded instructions. Capability inventory: The skill writes to persistent files like DISCOVERIES.md and PATTERNS.md, and generates new agent instruction files in the ~/.amplihack/.claude/agents/ directory (README.md, examples/agent-creation-example.md). Sanitization: Absent; the skill does not suggest any mechanisms for filtering, escaping, or validating the content extracted from user sessions.
Audit Metadata