rehab-therapy

Pass

Audited by Gen Agent Trust Hub on Mar 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill performs a structured audit of medical software codebases. It identifies clinical data models, recovery tracking metrics (ROM, strength, balance), and insurance billing workflows (8-minute rule, MIPS reporting). No unauthorized data access, network exfiltration, or prompt injection patterns were found.
  • [COMMAND_EXECUTION]: The skill uses the !command syntax via $ARGUMENTS to focus its analysis on specific parts of a codebase. This is a standard feature for developer tools on the target platform to allow user-driven scoping of automated tasks.
  • [DATA_EXFILTRATION]: While the skill mentions telemetry, it only logs execution metadata (success/failure status) to a local directory on the user's machine (~/.claude/projects/) for self-improvement purposes. No data is sent to external servers.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 23, 2026, 10:57 AM
Security Audit — agent-trust-hub — rehab-therapy