trigger-tasks

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • Indirect Prompt Injection (SAFE): The skill provides patterns for processing external data via task payloads. It mitigates injection risks by promoting the use of zod for strict schema validation. Mandatory Evidence Chain: 1. Ingestion points: Task payloads in run functions; 2. Boundary markers: Zod schemas in schemaTask; 3. Capability inventory: Background task execution; 4. Sanitization: Examples demonstrate validation of incoming data.
  • Unverifiable Dependencies & Remote Code Execution (SAFE): The skill utilizes official SDKs and standard libraries, with no indicators of unauthorized code execution or risky downloads.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 04:47 PM