memory

Pass

Audited by Gen Agent Trust Hub on Jun 28, 2026

Risk Level: SAFE
Full Analysis
  • [Local Workspace Interaction]: The skill records information into a file located at .agents/workspace/memory.md. This behavior is consistent with the skill's described purpose of maintaining a local interaction history within the project environment.
  • [Indirect Prompt Injection Surface]: The skill records user-supplied interaction data into a persistent log. This introduces a potential surface for indirect prompt injection if the logs are later retrieved and interpreted as instructions by an AI agent.
  • Ingestion points: User-agent interaction details as specified in SKILL.md.
  • Boundary markers: The suggested formatting lacks explicit boundary markers or delimiters to separate user data from system context.
  • Capability inventory: File writing capabilities within the local workspace environment.
  • Sanitization: User input is recorded directly to the file without specific escaping or filtering steps.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 28, 2026, 06:45 AM
Security Audit — agent-trust-hub — memory