Test Cloud admin guide

Context Grounding is accessible to all tenants and organizations.

For data residency details, refer to Data residency.

No, Context Grounding is available in both cloud and on-premises deployments.

Context Grounding charges for searches or RAG as it is executed through its supported UiPath product surfaces. For details, refer to Context Grounding licensing.

Context Grounding provides two services:

Context Grounding does provide a RAG service at runtime for UiPath GenAI experiences. However, it also provides a managed vector database as a service to help manage the data used at runtime. This guarantees a high-quality search and generated results.

You can use Context Grounding through UiPath GenAI Activities, Autopilot for Everyone, and Agents.

Context Grounding currently works the following data formats: PDF, JSON, CSV, DOCX, TXT, XLS.

To leverage Context Grounding, you need to import data into UiPath Orchestrator storage buckets (via direct upload, Studio activity, or API).

You can then use Context Grounding activities to ingest and index, and manage the queried data to ensure highly relevant results.

The limit of data you can use to ground your prompts is based on the model context window token size limits. Refer to the model you are using to execute the RAG (e.g., in GenAI activities) to determine potential token limit thresholds.

No, but it does significantly reduce the likelihood of hallucinations because generations are based on information queried from user-provided data. By default, Context Grounding provides a citation, or proof of knowledge, from which the generation was based. This means you can verify and validate the source. When Context Grounding isn't able to find a highly confident corresponding answer in the provided data, it does not try to make up answers. Instead, it generates a response such as: "An answer could not be found".

Yes. You can configure your own LLMs through the AI Trust Layer. The LLM configurations tab allows you to replace models or UiPath-managed subscriptions with your own models. Context Grounding also supports OpenAI v1 APIs, enabling compatibility with the latest OpenAI specifications and access to newer models.

In addition, within UiPath GenAI activities, you can select which LLM to use for executing the RAG process from those available in the LLM Gateway. UiPath manages ingestion and retrieval strategies to optimize results.

Context Grounding is tenant-scoped. We support folder-level authorization in Orchestrator buckets, and Context Grounding leverages existing authentication and Automation Ops policies.

All data shared with UiPath is treated with standard enterprise compliance, encryption, and security standards.

Context Grounding is part of the AI Trust Layer, which means your data is never stored outside of UiPath, nor is it used to train third-party models.

Context Grounding is tenant-scoped and takes advantage of existing RBAC and AuthZ policies in UiPath, in addition to encrypting data at rest and in transit.

Because it is tenant-scoped, no data is shared across indices within the same tenant or across tenants.

No. Files are held on UiPath infrastructure only for the duration of the ingestion run. After ingestion completes, raw files are not retained.

When you start an ingestion, Context Grounding fetches each file from the data source using the credentials you authorized through the configured Integration Services connection. No credentials are stored by Context Grounding.

Two short-lived copies exist while the job runs: a working copy on the processing node's local disk, and an internal staging copy used by the document-processing pipeline. Both are scoped to the active ingestion job and to your tenant, with no cross-tenant or cross-job access.

After ingestion completes, only the derived representation needed to power search and retrieval is retained: extracted text passages, embeddings, and structured metadata, bound to your dataset. Deleting the dataset removes these derived artifacts.

When a file is deleted from the data source, the next ingestion run — whether triggered manually or through a scheduled run — cascades the deletion to the corresponding index data.