skills/tonone-ai/tonone/brace-triage/Gen Agent Trust Hub

brace-triage

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill designs a triage system that ingests untrusted data from external users (support tickets) and uses it to perform automated actions like tagging and routing.
  • Ingestion points: Step 4 (Automation Rules) specifies that the agent should "Auto-tag" and route tickets based on "subject line keywords or form field selections".
  • Boundary markers: The instructions do not define delimiters or specific "ignore embedded instructions" warnings for the agent when processing these inputs.
  • Capability inventory: The skill's frontmatter allows high-capability tools such as Bash, WebFetch, and Read, which could be exploited if an attacker successfully injects instructions through a ticket.
  • Sanitization: There are no defined steps for sanitizing, validating, or escaping the external ticket content before the agent processes it for triage decisions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 02:57 AM
Security Audit — agent-trust-hub — brace-triage