khe-poteto-pyramid
Pass
Audited by Gen Agent Trust Hub on Nov 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The
sweepworkflow processes external data from the repository's history, which may contain untrusted content. - Ingestion points: The skill reads data from
git log,gh pr view --comments, and repository comments viagit grepas specified in thesweepjob withinSKILL.md. - Boundary markers: The instructions lack explicit boundary markers or directives for the agent to ignore potential instructions embedded within the ingested historical data.
- Capability inventory: The agent is empowered to refactor the codebase, modify dependency files, update linting configurations (such as
eslint.config.js), and execute shell commands defined in theEnforced by:fields of the generated records. - Sanitization: The skill does not provide mechanisms for sanitizing or validating the ingested history data before it influences the agent's logic for creating new project rules and automated checks.
Audit Metadata