release
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill reads and processes local files such as
package.json,AGENTS.md, and GitHub Actions workflows, and merges remote changes fromorigin/mainbefore executing build and verification commands. - Ingestion points: File reads of
package.jsonandAGENTS.md(Section 1), andgit mergeoperations fromorigin/main(Section 2). - Boundary markers: None identified; the agent is instructed to read the contents directly to determine the release state.
- Capability inventory: Subprocess execution for
git,bun,npm,gh, andnodecommands, as well as execution of the built CLI tool (node dist/cli.js) and its verification commands (jev-cli doctor) (Sections 1, 2, 4, 5, and 6). - Sanitization: No explicit sanitization or validation of the contents of the read files is described before they are used to influence the release process or versioning logic.
Audit Metadata