signal-interpreter

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted external data from signal providers and enrichment services, creating a surface for indirect prompt injection.
  • Ingestion points: Untrusted content is ingested via parameters such as event_details, enrichment_data, account_memory, and intent_corroborating_signals defined in SKILL.md.
  • Boundary markers: The instructions do not provide specific delimiters or ignore-instructions logic to isolate these external inputs from the system prompt.
  • Capability inventory: The skill is designed purely for data interpretation and lacks capabilities for file system access, shell command execution, or network operations.
  • Sanitization: No methods for input validation, escaping, or content filtering are specified to prevent malicious data from influencing the agent's logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 07:26 PM
Security Audit — agent-trust-hub — signal-interpreter