Salesforce

Salesforce Agentforce Audit Logging & AI Evidence

Provable agent actions inside Salesforce

Agentforce agents take CRM actions, call tools, and escalate to humans. Sigigo witnesses those decision points with the same evidence schema — without replacing Salesforce audit trails.

Where Sigigo sits

Sigigo is evidence infrastructure alongside Salesforce — not a replacement for Setup Audit Trail or Event Monitoring. We certify the AI-specific chain: what the agent saw, which tools fired, and what changed in customer records.

Typical attachment points

  • Before/after Einstein / Agentforce model calls (commitment over prompt context)
  • Flow and Apex boundaries when agents invoke actions
  • Human approval steps — origin-signed overrides for Article 14-style oversight
  • Exportable audit packages for customer security reviews

Data residency

CRM payloads stay in Salesforce or your vault. Sigigo stores proofs and metadata under region-locked tenant policy for EU enterprise deployments.

Evidence events

Canonical event types emitted through this integration path:

  • ai_act.inference_started / completed (agent turn)
  • ai_act.action_executed (Flow, Apex, external tools)
  • ai_act.human_review_* (escalation / approval)
  • data_act.* when agents trigger data access or sharing

Related

Integration availability may vary by tenant, region, and deployment model. This page describes product architecture — not a guarantee of feature availability. See compliance disclaimers and contact us for early access.