code-understanding

Pass

Audited by Gen Agent Trust Hub on Jun 7, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes several internal utility scripts located in the libexec/ directory (e.g., mantishack-normalize-context-map, mantishack-coverage-summary) to normalize analysis data and generate coverage reports for the validation pipeline.
  • [REMOTE_CODE_EXECUTION]: Provides an optional "Runtime Probe" capability designed to execute a target's binary within a controlled sandbox environment (sandbox(observe=True)) to corroborate static analysis with runtime observations of file and network activity.
  • [SAFE]: Implements a robust "Adversarial Reasoning" framework through mandatory gates (GATE-U1 through GATE-U5) which require all analytical claims to be backed by direct code evidence and quoted file paths, effectively mitigating risks from indirect prompt injection that might be present in analyzed codebases.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 7, 2026, 01:29 PM