shape
Pass
Audited by Gen Agent Trust Hub on Nov 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's workflow involves reading 'facts' from external sources such as files and command outputs to map out a decision tree. These outputs are used to generate questions and summary records that are written back to the project (e.g., Pull Request bodies or Issue descriptions), which constitutes a vulnerability surface for indirect prompt injection.
- Ingestion points: The skill explicitly instructs the agent to find 'facts' from files and command outputs (SKILL.md, Procedure step 2).
- Boundary markers: There are no instructions to use delimiters or to disregard potentially malicious instructions embedded in the 'facts' or files being read.
- Capability inventory: The skill possesses the capability to write to pull request bodies, issues, and project plans, and to coordinate with other tools (SKILL.md, Records section).
- Sanitization: The procedure does not specify any sanitization or validation of the data retrieved from external sources before it is used to influence the agent's interactive questions or its final summary output.
Audit Metadata