open-change-request
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting untrusted data from external sources to populate pull request fields.
- Ingestion points: External issue titles from the tracker and branch metadata are retrieved and processed as described in
SKILL.mdto form the PR title and body. - Boundary markers: Absent; the instructions do not specify delimiters or warnings to ignore instructions that might be embedded within the issue titles.
- Capability inventory: The skill is capable of performing git push operations, network API calls to create pull/merge requests, and network API calls to update issue tracker states.
- Sanitization: Absent; there are no explicit instructions to sanitize or validate the content retrieved from the issue tracker before it is interpolated into the pull request template.
Audit Metadata