prepare
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- Local Command Execution: The skill utilizes shell commands through tools like
gh,git,python3, anduvto automate release preparation. For instance, it executes a pre-flight check usingpython3 -m setup_preflightand performs archive comparisons viauv run. These actions are standard for repository management and are scoped to the local environment. - Indirect Prompt Injection Surface: The skill processes external data such as pull request titles, issue bodies, and changelogs. While this presents a potential injection surface, the skill includes specific directives to treat this content as data rather than instructions and provides a framework for flagging suspicious input, which follows security best practices.
- Human-in-the-Loop Confirmation: Every state-changing action, including creating GitHub issues or drafting pull requests, requires explicit confirmation from the Release Manager. This ensures that the agent does not perform autonomous merges or closures, maintaining user control over the release process.
- Temporary File Usage: The skill uses temporary files (e.g., in
/tmp/) to stage issue bodies. This is a common and safe pattern to handle multi-line data during automation while avoiding shell-quoting issues. - External Resource Access: The skill references official documentation and licenses from the Apache Software Foundation and interacts with the project's GitHub repository. These interactions are consistent with the skill's purpose and the trusted nature of the project's own infrastructure.
Audit Metadata