data-partitioner

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [General] (SAFE): No evidence of malicious behavior, prompt injection, or data exfiltration was found. The skill consists only of documentation and metadata.
  • [Indirect Prompt Injection] (LOW): The skill operates in a data engineering context (ETL/Spark) where it may process untrusted data using powerful tools (Bash, Write). This creates a structural attack surface, although no vulnerability is present in the skill itself. 1. Ingestion points: Data pipeline tasks and requests. 2. Boundary markers: None defined. 3. Capability inventory: Bash, Write, Edit, Read, Grep. 4. Sanitization: Not specified.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:36 PM