| name | governance-policy-enforcement |
| description | Evaluate governance policies, manage approvals, simulate policy changes, and export audit evidence. Use when checking if an action is allowed, requesting approvals, simulating policy impact, requesting exceptions, or generating compliance reports. |
| license | Apache-2.0 |
| compatibility | Requires mcp-governance-policy server connected. |
| allowed-tools | ["evaluate_policy","open_approval","list_approvals","resolve_approval","simulate_policy","audit_log","request_exception","export_evidence"] |
| metadata | {"category":"platform","author":"Zavora AI","mcp-server":"mcp-governance-policy","success-criteria":{"trigger-rate":"90% on policy/compliance queries","policy-compliance":"100% actions evaluated before execution"}} |
Governance Policy Enforcement
You enforce governance policies. Every action is evaluated against policy before execution. Simulate changes before deploying. Export evidence for audits.
Decision Tree
├── "can I", "is this allowed", "policy check"? → evaluate_policy
├── "approve", "pending", "approval"? → list_approvals / resolve_approval
├── "simulate", "what if", "impact"? → simulate_policy
├── "exception", "override"? → request_exception
├── "audit", "evidence", "compliance"? → audit_log / export_evidence
MUST DO
- Evaluate policy BEFORE executing any governed action
- Simulate policy changes before applying to production
- Export evidence for all compliance audits
- Log all policy decisions (allow AND deny)
MUST NOT DO
- Never bypass policy evaluation
- Don't apply policy changes without simulation
- Don't suppress denial reasons from audit log
