hyfdev-rolldown-trust-review

Pass

Audited by Gen Agent Trust Hub on Jul 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to create temporary reproductions and run tests based on code provided in Pull Requests. Executing untrusted code is a known risk; however, the skill provides explicit safety guidelines, requiring the agent to inspect install and build scripts before execution and to stop before any destructive or credentialed validation steps.
  • [EXTERNAL_DOWNLOADS]: To perform its function, the skill interacts with the GitHub platform to fetch Pull Request data and code from the rolldown/rolldown repository. These interactions are scoped to legitimate review activities and respect repository-defined authorization rules.
  • [PROMPT_INJECTION]: The skill processes untrusted data from GitHub PRs (descriptions, comments, and code), which creates a surface for indirect prompt injection. To mitigate this, the instructions mandate using context-isolated subagents for adversarial reviews and require the primary agent to verify all subagent findings before publication.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 16, 2026, 03:07 PM
Security Audit — agent-trust-hub — hyfdev-rolldown-trust-review