agent-genesis
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by instructing agents to incorporate knowledge from the
.agents/agent_md/directory into their operational rules.\n- Ingestion points: Files within the.agents/agent_md/directory and the project codebase (SKILL.md).\n- Boundary markers: None identified. There are no instructions to isolate project data from the agent's core instructions.\n- Capability inventory: The agent can read codebase files and write to the persistent memory directory (SKILL.md).\n- Sanitization: None identified. The protocol encourages adopting discovered patterns without verification.
Audit Metadata