handover

Pass

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates as intended for conversation management and project handoffs.
  • [COMMAND_EXECUTION]: The skill directs the agent to use a 'Write' tool to save documentation to the local filesystem. This is a standard and expected operation for a file-generation utility.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data from the conversation history. 1. Ingestion points: Conversation history read by the agent (SKILL.md). 2. Boundary markers: Uses a specific Markdown template for output structure. 3. Capability inventory: Uses the 'Write' tool to persist data (SKILL.md). 4. Sanitization: Implements 'Writing Guidelines' and a 'Self-Verification' checklist requiring the agent to distinguish between facts and inferences, which mitigates the risk of malicious instructions being inadvertently adopted or propagated.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 7, 2026, 09:22 AM
Security Audit — agent-trust-hub — handover