handoff-update
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill implements a workflow that reads and processes data from an existing handoff document, which acts as an untrusted input surface.
- Ingestion points: The agent reads a file resolved from user-provided arguments or conversation history in Step 2.
- Boundary markers: The instructions lack delimiters or explicit directives to ignore instructions that might be embedded within the handoff file being updated.
- Capability inventory: The skill has permissions to read and write to the local filesystem to update the documentation.
- Sanitization: There is no mentioned validation or filtering process for the content read from the file before it is merged and rewritten in Step 3.
- [EXTERNAL_DOWNLOADS]: The skill mentions a template source from a GitHub repository (github.com/packmind/packmind). This is a well-known service and the reference is for attribution purposes only; the skill does not automate any downloads or execution of remote code from this source.
Audit Metadata