remember

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFE
Full Analysis
  • Persistent Configuration Access: The skill is designed to modify files in the ~/.deepagents/ directory to store best practices and coding conventions. This mechanism ensures that learnings are preserved across different conversations and projects.
  • Instruction Evolution Surface: This skill incorporates an 'Indirect Prompt Injection' surface as it ingests data from conversation history to update its future instructions. (1) Ingestion points: Conversation history (SKILL.md). (2) Boundary markers: None identified in the skill instructions. (3) Capability inventory: The skill utilizes edit_file and write_file to modify agent memory and create new skill instructions. (4) Sanitization: No explicit sanitization or validation of the ingested conversation data is defined before it is committed to persistent storage.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 02:43 AM
Security Audit — agent-trust-hub — remember