The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external CRM systems, which serves as a surface for untrusted data. * Ingestion points: CRM data is extracted via the crm_accounts and crm_contacts models defined in the infra/models folder. * Boundary markers: The skill implements strict zod schema validation for all workflow inputs and outputs in account-enrichment.ts and contact-linkedin-enrichment.ts. * Capability inventory: The skill utilizes crm.updateRecords for data writing and the LinkedIn connector for external API calls. * Sanitization: Input data is validated against expected types and structures before processing.
[COMMAND_EXECUTION]: The documentation and skill instructions provide standard setup, installation, and verification commands using the cargo-ai CLI and Node.js to ensure pipeline configuration integrity.

crm-enrichment