skills/versusjl/devin-blueprint/pr/Gen Agent Trust Hub

pr

Pass

Audited by Gen Agent Trust Hub on Jul 2, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: Uses the git and gh (GitHub CLI) tools to inspect local changes, commit code, and push to remote repositories.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing untrusted repository data into pull request descriptions.
  • Ingestion points: Repository metadata and code changes are read via git status and git diff (SKILL.md).
  • Boundary markers: No explicit delimiters are used to wrap the repository content before processing.
  • Capability inventory: Includes the ability to commit changes, push to remote servers, and create pull requests via network-enabled tools.
  • Sanitization: There is no automated sanitization for content ingested from the repository diffs, though the instructions explicitly warn the agent not to stage secrets.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 2, 2026, 11:05 PM
Security Audit — agent-trust-hub — pr