crm-integration-skill

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The documentation recommends the installation of well-known and standard Python libraries, including httpx, pyjwt, hubspot-api-client, and python-dotenv, from official package registries to implement the CRM integration patterns described in the skill.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external CRM platforms (leads, contacts, deals) through various search and retrieval operations, which constitutes a standard surface for indirect prompt injection.
  • Ingestion points: CRM API response data from Close CRM, HubSpot, and Salesforce, as well as enrichment data from Clay and Apollo (referenced in SKILL.md and reference/clay-enrichment-patterns.md).
  • Boundary markers: No specific boundary markers or 'ignore' instructions for embedded content are defined in the provided code snippets.
  • Capability inventory: The skill includes capabilities for network communication (via httpx and SDKs) and file system writes for recording execution metrics (the 'Sidecar Emit' directive in SKILL.md).
  • Sanitization: The provided templates do not include specific sanitization logic for data retrieved from external CRM APIs before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 09:52 PM