Governance
role-based access Policy Enforcer
Posted May 13, 2026
Apply guardrails-backed role and attribute policies to every tool call before execution.
Every skill invocation passes through guardrails & approvals evaluation. Deny decisions are logged and the workflow halted, no silent failures.
guardrails
Policy engine
<5ms
Eval latency
role-based access + attribute-based access
Policy types
100%
Audit events
Details
The essentials for role-based access Policy Enforcer: everything published so far.
Category
Governance
Status
Production
Version
v2.0.0
Posted
May 13, 2026
Skill manifest
The exact instructions this skill ships with, shown in full. Nothing hidden.
--- id: rbac-policy-enforcer version: v2.0.0 display_name: role-based access Policy Enforcer category: governance category_label: Governance status: Production gateway: Secure Connector tags: ["guardrails", "Rego", "role-based access", "attribute-based access", "Policy as Code"] --- # role-based access Policy Enforcer > Apply guardrails-backed role and attribute policies to every tool call before execution. Every skill invocation passes through guardrails & approvals evaluation. Deny decisions are logged and the workflow halted, no silent failures. ## Stats - **Policy engine:** guardrails - **Eval latency:** <5ms - **Policy types:** role-based access + attribute-based access - **Audit events:** 100% ## Supported operations ### Role check Assert caller holds required role before allowing skill execution. ### Attribute check Evaluate attribute-based access conditions on resource attributes, env context and user claims. ### Policy-as-code Author Rego policies version-controlled alongside your workflow definitions. ### Deny logging Every deny decision written to immutable audit trail with policy rule reference. ## How a call flows 1. Skill invocation (caller + context) 2. guardrails evaluator (Rego policy check) 3. Allow / Deny (decision in <5ms) 4. Workflow continues or halts on deny 5. Audit trail (decision + rule logged) ## Input schema | Field | Type | Required | Description | | --- | --- | --- | --- | | `resource` | `string` | yes | skill or connector being accessed | | `caller_id` | `string` | yes | user or service identity | | `caller_roles` | `string[]` | yes | roles from identity token | | `attributes` | `object` | no | resource and environment attributes for attribute-based access evaluation | | `policy_package` | `string` | no | Rego package name to evaluate (default "agento.authz") | | `context` | `object` | no | additional context passed to policy (time, IP, team, env) | ## Audit and evidence ### Every decision logged Allow and deny decisions written with caller identity, resource, policy rule and timestamp. ### Policy version pinned Policy package version pinned to workflow version, policy changes require new deployment. ### Sub-5ms evaluation guardrails evaluation adds under 5ms to skill invocation latency, no workflow impact. ### Rego audit trail Exact Rego rule that fired stored alongside decision for regulator review. ## Example use case: Finance Team Access Control 1. **Workflow triggered.** AP automation workflow triggered by new invoice in shared mailbox. 2. **role-based access check.** Policy Enforcer evaluates: does caller hold "finance-ap" role? Deny if not. 3. **attribute-based access check.** For invoices over $50k, additional "finance-senior" attribute required. 4. **Policy logged.** Exact Rego rule and decision written to immutable audit trail. 5. **Workflow proceeds.** On allow, workflow continues to Slack approval and SAP posting steps. --- *Generated from `src/app/data/skills.ts` (slug: `rbac-policy-enforcer`). Do not hand-edit. Re-run `npm run generate:skill-manifests` after changing the source data.*
Reviews
Reviews from teams using role-based access Policy Enforcer.
No reviews yet
Be the first to share how role-based access Policy Enforcer works for your team.
Discussion(0)
No comments yet. Be the first.