tag-release
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill relies on external data from the repository, such as existing tag names and release status, which could be manipulated to influence agent behavior.\n
- Ingestion points: Data is fetched via
git ls-remote,gh release list, and GitHub API calls for commit status and check runs as defined in SKILL.md.\n - Boundary markers: The instructions do not specify any delimiters or safety prompts to isolate external data from the agent's core instructions.\n
- Capability inventory: The skill possesses the ability to execute
git tag,git push, andgh release create, enabling it to commit changes to the remote repository.\n - Sanitization: There are no explicit steps in the instructions to sanitize or validate the format of version strings or release content retrieved from the environment.
Audit Metadata