expand-tasks

Pass

Audited by Gen Agent Trust Hub on Aug 5, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing external research data. 1. Ingestion points: research agent outputs and tasks.json. 2. Boundary markers: none specified to delimit research content. 3. Capability inventory: uses Write, Bash, and Task tools. 4. Sanitization: none mentioned before writing back to tasks.json.
  • [COMMAND_EXECUTION]: The skill executes local Python scripts to handle deterministic operations and update task state.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 5, 2026, 11:14 AM
Security Audit — agent-trust-hub — expand-tasks