ultrawork

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill acts as a router, directing the agent to a local workflow file for instructions. No suspicious code, network activity, or obfuscation was found.
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and follow a local file .agents/workflows/ultrawork.md. This represents a surface for instructions to enter the context, though it is standard for workflow management.
  • Ingestion points: .agents/workflows/ultrawork.md (referenced in SKILL.md).
  • Boundary markers: Absent.
  • Capability inventory: The skill frontmatter includes disable-model-invocation: true, which limits the agent's tool-calling capability within this skill context.
  • Sanitization: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 09:30 PM
Security Audit — agent-trust-hub — ultrawork