handoff
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the workspace and conversation history to generate a handoff document that provides instructions and context for a subsequent agent session.
- Ingestion points: The skill reads the current conversation history, workspace file system, git state (branches, worktree paths, file changes), and verification results from the local environment.
- Boundary markers: The instructions do not specify structured delimiters or safety boundaries for the untrusted content being summarized into the document.
- Capability inventory: The skill utilizes file-writing capabilities to save the generated handoff document to the
.tmp/handoff/directory within the workspace. - Sanitization: The skill includes an explicit instruction for the agent to redact sensitive information such as API keys, passwords, and PII before generating the artifact, which mitigates simple credential exposure.
Audit Metadata