electrobun-milady
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes an integration where an agent's Pull Request (PR) body is machine-parsed by a CI/CD pipeline to perform automated repository actions.\n
- Ingestion points: The 'Follow-up' section of the PR body template in
SKILL.mdis designated for automatic conversion into GitHub issues.\n - Boundary markers: The skill provides a Markdown template but does not define strict delimiters or warnings to prevent the parsing system from executing instructions embedded within the follow-up items.\n
- Capability inventory: The documented
review-postprocessandauto-mergejobs in the project's workflow possess the ability to create issues, apply labels, and merge or close PRs.\n - Sanitization: There is no evidence of content sanitization or validation for the text being auto-converted from PR bullets into issues.
Audit Metadata