handover
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest data from external sources such as background job logs, project status files, and previous handover notes. This creates a surface for indirect prompt injection if those files contain malicious instructions intended to influence the agent's behavior.
- Ingestion points:
git status, background job logs, project state files (backlogs, status boards), and inherited handover reports (SKILL.md Step 1). - Boundary markers: No specific delimiters or safety wrappers are suggested for the content read from these files.
- Capability inventory: The skill is intended to be used by agents with the ability to read the file system and write documentation (e.g.,
HANDOVER.md). - Sanitization: The skill provides a negative constraint, explicitly stating "Nothing secret goes in it: no keys, tokens, passwords or connection strings."
Audit Metadata