therapy-outcomes

Pass

Audited by Gen Agent Trust Hub on Mar 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill performs legitimate analysis of therapy outcome metrics based on provided codebase data.
  • [DATA_EXPOSURE]: The skill accesses the ~/.claude/projects/ directory to manage execution telemetry. This is a local operation for project state management within the intended platform environment.
  • [PROMPT_INJECTION]: The skill processes external data via $ARGUMENTS and codebase files. Ingestion points: User-provided arguments and the project codebase. Boundary markers: None present. Capability inventory: Read access to project files; write access to the docs/ directory and project metadata. Sanitization: No explicit sanitization of ingested content. Risk: Indirect prompt injection is possible if the processed data contains malicious instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 23, 2026, 10:57 AM
Security Audit — agent-trust-hub — therapy-outcomes