- はじめに
- ベスト プラクティス
- 機能
- スキル
- CLI
- 例
- 例の概要
- レシピ: エージェントを構築する
- レシピ: RPA で自動化する
- レシピ: Web アプリへのデータ入力
- レシピ: デスクトップ アプリからデータを抽出します
- レシピ: ワークフローをリファクタリングしてテストする
- レシピ: REFramework を使用したキューベースのプロセス
- レシピ: Maestro Flow を構築する
- Recipe: build a Maestro Case
- レシピ: コード化されたアプリと API ワークフロー
- Recipe: build a coded function
- レシピ: Test Manager でリリースを検証する
- レシピ: IXP を使用してドキュメントからデータを抽出する
- Advanced (カスタム)
- ヘルプ
Governance policies for authoring, deploying, and diagnosing product feature restrictions, tool-use permissions, and compliance posture across a UiPath tenant.
A coding agent can author and deploy governance policies across your UiPath tenant from a plain-language description of the rule you want enforced, then diagnose why a policy isn't taking effect.
機能
Governance policies work on two layers, and the agent classifies your request into the right one before authoring anything:
- Product policies — block, restrict, or enforce specific features in Studio, StudioX, Assistant, Robot, AI Trust Layer, or Agent Builder. Deployed to a user, a group, or a whole tenant, with tenant-level settings overridden by group, and group by user.
- Access policies — allow or deny one workflow from invoking another as a tool, for example an agent calling a Maestro process, a Flow, an RPA workflow, an API workflow, or a case. Rules can be gated by tag, by caller, or by the acting user or group.
The agent can also check a tenant's compliance posture against a standard such as ISO 42001 or ISO 27001 and configure the recommended settings across products in one operation, always showing you the plan before applying anything.
Diagnosing policy behavior
Beyond authoring, the agent can investigate why a policy isn't taking effect, work through deployment precedence when user, group, and tenant settings conflict, and explain why a specific tool invocation was blocked.
その背後にあるスキル
This work is powered by the uipath-governance skill, which authors, deploys, and diagnoses policies across both layers, and checks compliance posture against supported standards. You can see it in the Skills catalog.
一般的な開始プロンプトです
最初の要求は次のようになります。
"Block Agent Builder from using any model outside our approved list, for the whole tenant."
The agent classifies the request, shows you the policy it intends to create, and asks for confirmation before deploying it.
その他のプロンプトを試す
- Deny any agent from invoking an RPA workflow tagged "finance-restricted" unless the caller is in the Finance Automation group.
- Why isn't the Studio policy I deployed to the Sales group taking effect for a specific user?
- Check our tenant's compliance posture against ISO 27001 and show me what's missing.
ウェビナーに参加
For the full picture of governance policies, see the Automation Ops governance documentation.