skills/builderio/agent-native/brain/Gen Agent Trust Hub

brain

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues were detected in the analyzed files. The skill and internal runbook describe a well-architected system focusing on data privacy and security.
  • Data Sanitization: The skill includes comprehensive logic to strip sensitive information—including PII, recruiting data, and API-key-shaped strings—from transcripts before they are stored in the database.
  • Credential Management: Credentials like Slack tokens are resolved using a secure hierarchy (workspace connections and registered vaults) rather than being hardcoded or read directly from insecure environment variables.
  • Access Control: The system implements tiered access, requiring specific 'editor' permissions for retrieving raw, unredacted content from sources.
  • Indirect Prompt Injection Mitigation: The use of deterministic sanitization and SQL-based text matching for search results reduces the attack surface for indirect prompt injection compared to unvalidated vector-based retrieval systems.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 04:51 AM
Security Audit — agent-trust-hub — brain