crm-integration
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest user-provided CRM requirements to generate integration blueprints, which introduces a potential surface for indirect prompt injection if the user provides malicious instructions within those requirements.
- Ingestion points: User descriptions of CRM setups, deal stages, and business processes are processed by the agent to create documentation as described in
SKILL.md. - Boundary markers: The instructions do not define specific delimiters or guardrails to separate user-provided data from agent instructions during the blueprint generation process.
- Capability inventory: The skill includes a local validation script
scripts/check-output.pythat reads files to check for specific markdown headers using standard file I/O. - Sanitization: There is no explicit sanitization or filtering of the input data before it is processed or used to generate deliverables.
Audit Metadata