clade-data-handling

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, persistence mechanisms, or unauthorized network operations were detected.
  • [PROMPT_INJECTION]: The skill defines logic for processing conversation messages and text, which creates a surface for indirect prompt injection.
  • Ingestion points: Processes 'messages' and 'text' strings as described in the TypeScript snippets in SKILL.md.
  • Boundary markers: No explicit boundary delimiters are present in the code examples to isolate untrusted data.
  • Capability inventory: The skill is configured to use 'Read', 'Write', and 'Edit' tools.
  • Sanitization: The skill provides regex-based PII redaction for emails, phone numbers, SSNs, and credit cards as a security measure.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 01:05 AM
Security Audit — agent-trust-hub — clade-data-handling