simplify

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data in the form of code diffs and repository-level configuration files which could contain malicious instructions.
  • Ingestion points: The agent reads output from git diff, staged changes, and local files like AGENTS.md to identify review targets and standards (SKILL.md, Phase 1 and Phase 2).
  • Boundary markers: The instructions do not define clear boundaries or delimiters to isolate ingested code from the agent's instructions, nor do they include warnings to ignore instructions embedded in the code.
  • Capability inventory: The agent is authorized to modify files to apply fixes (SKILL.md, Phase 4) and execute shell commands for linting, building, and testing (SKILL.md, Phase 5).
  • Sanitization: The skill does not implement validation or sanitization of the content extracted from diffs before processing it through the review sub-agents.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 08:16 AM
Security Audit — agent-trust-hub — simplify