release-prep
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to read and process untrusted data from the repository, such as unreleased changelog entries, issue descriptions, and pull request titles, to generate release notes and documentation. Malicious instructions embedded in these external sources could potentially influence the agent's summarization or documentation behavior.
- Ingestion points: Reads project files (package.json, go.mod, VERSION), changelogs, README.md, and reviews open issues/PRs.
- Boundary markers: Absent.
- Capability inventory: Performs file system writes (documentation and version files) and executes shell commands (tests and linting).
- Sanitization: None identified; the agent processes the content as plain text.
- [COMMAND_EXECUTION]: The skill is instructed to execute the project's internal test and linting suites. This involves running code defined within the repository's own configuration files. While this is the intended primary purpose of the skill, it relies on the integrity of the local environment's test scripts.
Audit Metadata