Governance RCA

Purpose

Governance RCA produces bounded attribution evidence for material governance outcome differences.

Use Governance RCA when a customer needs to understand whether a candidate policy change is responsible for a changed runtime outcome.

When To Use It

Use Governance RCA when:

• A valid replay_reference is available.
• Candidate policy setup is available.
• Comparison or simulation output indicates a difference that needs bounded attribution.
• The customer wants runtime evidence to support review, remediation, approval, rollback, or release decisions.

Request Boundary

Governance RCA requires:

• replay_reference.
• candidate_policy_setup.

Governance RCA composes replay, simulation, comparison, and bounded governance attribution. It does not bypass those runtime paths.

Response Evidence

A successful Governance RCA response may include:

• Status.
• replay_reference.
• Baseline outcome.
• Simulated outcome.
• Comparison summary.
• root_cause.
• Bounded errors.

The root_cause object may indicate:

• status: "attributed" when comparison reports a material governance outcome difference.
• status: "no_material_difference" when comparison reports no material governance outcome difference.
• Responsible governance change identity where attribution is available.
• Attribution type.

Customer Responsibilities

Customers should:

• Store RCA evidence with related replay, simulation, comparison, governance, policy, review, and approval records.
• Decide remediation, approval, promotion, rollback, or governance evolution outcomes in customer-owned systems.
• Treat RCA as bounded runtime evidence, not as an automatic decision.

Boundary

Governance RCA is limited to bounded governance attribution. It is not enterprise RCA, infrastructure RCA, organizational RCA, clinical RCA, financial RCA, or an explanation engine.

Next Step

Use Runtime API Reference for the customer-facing API surface summary, then consult foxcommand-runtime for exact request and response contracts.