math-review

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions to execute several local system commands including git for repository synchronization, pytest for running mathematical unit tests and benchmarks, and jupyter nbconvert for executing notebooks. These commands are necessary for its primary function of code review and are limited to local files.
  • [REMOTE_CODE_EXECUTION]: The skill uses jupyter nbconvert --execute derivation.ipynb to verify mathematical derivations. Although this command executes the code contained within the notebook, this is the intended and necessary functionality for symbolically verifying mathematical properties as defined in the skill's core workflow.
  • [INDIRECT_PROMPT_INJECTION]: The skill contains an inherent attack surface for indirect prompt injection because it processes and executes mathematical logic provided in external files.
  • Ingestion points: The skill analyzes and executes local files, specifically derivation.ipynb and files within the tests/math/ directory.
  • Boundary markers: None; the skill is designed to treat the repository content as the subject of mathematical analysis.
  • Capability inventory: The skill performs subprocess execution using pytest and jupyter as described in SKILL.md and modules/testing-strategies.md.
  • Sanitization: None; the skill relies on the user providing legitimate mathematical code for verification purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 01:34 AM