magpie-security-issue-import-from-pr
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [Secure Input Processing]: The skill provides explicit instructions to the agent to prevent shell injection. When processing attacker-controlled content like pull request titles, it directs the agent to write the content to a temporary file and use the GitHub CLI's file-loading flag (
-F title=@file) rather than inlining the data directly into a shell command. - [Prompt Injection Safeguards]: Because the skill ingests external data from public PRs, it includes a dedicated section warning the agent that this content should be treated strictly as data and never as instructions. It specifically instructs the agent to flag and ignore any directives found within external PR bodies or comments.
- [Least Privilege and Transparency]: The skill operates using the standard GitHub CLI (
gh), relying on the user's existing environment permissions. It requires explicit user confirmation before performing any write operations, providing a full preview of the proposed issue content, labels, and project board placement. - [Verified Source]: The skill is authored by a trusted organization (Apache) and interacts exclusively with official project infrastructure. It includes checks for configuration drift and authenticity of its own components to maintain environment integrity.
Audit Metadata