code-review

Pass

Audited by Gen Agent Trust Hub on Oct 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands using uv run, including running pytest for verification and the mp CLI for API reference checks.
  • [DYNAMIC_EXECUTION]: The procedure instructs the agent to run arbitrary Python code snippets via python -c to confirm behaviors observed in the Pull Request code. This involves executing logic that originates from the potentially untrusted source code being reviewed.
  • [INDIRECT_PROMPT_INJECTION]: The skill is exposed to indirect prompt injection by reading and processing data from Pull Request descriptions and external instruction files like CLAUDE.md and .github/instructions/.
  • Ingestion points: Pull Request descriptions, .github/instructions/, CLAUDE.md, and REVIEW.md (File: SKILL.md).
  • Boundary markers: Absent. No delimiters or instructions to ignore embedded commands are specified for the ingested content.
  • Capability inventory: The skill can execute shell commands and Python code via uv run (File: SKILL.md).
  • Sanitization: No sanitization or validation of the ingested content is performed before it is used to influence agent actions.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 12, 2026, 03:24 PM