communication-routing

Pass

Audited by Gen Agent Trust Hub on Mar 15, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted message payloads from external agents, creating an attack surface for indirect prompt injection.
  • Ingestion points: The message JSON input defined in SKILL.md is the primary entry point for untrusted data.
  • Boundary markers: There are no explicit delimiters or instructions provided to differentiate between data and instructions within the message payload.
  • Capability inventory: The skill is limited to message routing, audit logging, and delivery confirmation; it does not possess capabilities for file system access, network requests (beyond internal routing), or command execution.
  • Sanitization: Step 2 (Validate Routing) includes checks for source authorization, target existence, and payload format validity.
  • [NO_CODE]: The skill is entirely descriptive, consisting of markdown-based instructions and process workflows without any scripts, binaries, or shell commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 15, 2026, 04:55 AM