crm-integration

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns or high-risk vulnerabilities were identified. The skill correctly uses environment variables for API keys and employs well-known, legitimate libraries for CRM communications.- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface through the ingestion of external CRM data. 1. Ingestion points: Data enters the context via CRM search and retrieval tools listed in SKILL.md and enrichment polling in reference/clay-enrichment-patterns.md. 2. Boundary markers: No delimiters are used to isolate untrusted external content. 3. Capability inventory: The skill has network access via httpx and file-write capabilities for local telemetry. 4. Sanitization: External data is not sanitized or escaped before use in subsequent agent actions. This risk is inherent to the skill's primary function.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 09:56 AM
Security Audit — agent-trust-hub — crm-integration