ai-agent-tooling-and-hitl

Pass

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a technical reference for securing AI agents. It explicitly defines security best practices, such as classifying tools by reversibility, implementing human approval gates for irreversible actions, and enforcing side-effect budgets to prevent runaway agent loops.
  • [SAFE]: No malicious patterns, prompt injections, or unauthorized data access were detected. The included code snippets (Python, PHP, SQL, TypeScript) are illustrative examples for building security infrastructure rather than executable payloads.
  • [SAFE]: The skill incorporates defensive strategies against indirect prompt injection by recommending the use of concise tool observations and strict JSON schema validation for all tool inputs and outputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 15, 2026, 04:57 PM
Security Audit — agent-trust-hub — ai-agent-tooling-and-hitl