Palantir Foundry AIP Agents

Openlayer connects to your Palantir Foundry AIP Chatbots to ingest execution traces for online evaluations. AIP Chatbots were formerly called AIP Agents. The integration polls a Chatbot Studio session-log streaming dataset, groups exported rows into complete executions, and writes them to the Openlayer projects linked to your chatbots.

​

How it works

Once connected, Openlayer:

1. Discovers your chatbots — walks Compass folders for AIP Chatbot files, or lets you register a chatbot by agent RID or filesystem path.
2. Links Openlayer projects — enabling a chatbot links it to an Openlayer project and inference pipeline. By default, Openlayer creates an LLM project named Foundry AIP Agent - <name>.
3. Syncs executions — periodic or manual sync polls the session-log stream and ingests completed executions for enabled chatbots whose agent_rid or owning_rid matches.
4. Builds traces — Openlayer groups rows by traceId and maps chat messages, tool calls, outputs, and errors into structured traces.

Unknown chatbots are dropped rather than registered automatically. Incomplete traces remain pending until the stream includes a final_response or execution_error.

---

​

Prerequisites

Before connecting, you need:

• A Foundry enrollment with at least one AIP Chatbot.
• A bearer token or third-party application token with api:aip-agents-read, api:filesystem-read, and api:streams-read.
• An Org Admin-created Chatbot Studio session-log export using the Palantir JSON schema.
• An Openlayer workspace where you are an admin.

Your enrollment URL must use HTTPS and contain only the hostname, such as https://example.palantirfoundry.com, with no path.

​

Create the session-log export

As a Foundry Org Admin:

1. In Control Panel, select your organization.
2. Open Log observability settings and click Create log export.
3. Select the projects that contain the chatbots you want to monitor.
4. Choose the Palantir JSON schema and create the streaming dataset.
5. Copy the dataset RID. Allow up to five minutes for rows to appear.

Use Palantir JSON rather than OpenTelemetry protobuf for this integration.

---

​

Set up the integration

​

Step 1: Connect your enrollment

1. In Openlayer, go to Settings → Integrations.
2. Select Palantir Foundry AIP Agents.
3. Under Connect, enter:
   • Enrollment URL — your HTTPS Foundry enrollment hostname with no path.
   • Bearer token — a token with all three required scopes.
   • Session-log dataset RID (optional) — the RID from your Org Admin’s log export.
   • Stream branch — the dataset branch, which defaults to master.
4. Click Connect.

Openlayer verifies the hostname and token before saving the connection.

​

Step 2: Discover or register chatbots

Click Discover agents to walk the Compass folders visible to your token for AIP Chatbot files. You can also click Register agent and provide either:

• The agent RID from the Chatbot Studio URL.
• A Foundry filesystem path, such as /Org/Project/My Agent.

Discovery only finds Compass files with the AIP_AGENTS_AGENT type.

​

Step 3: Enable chatbots

Click Enable for each chatbot you want to monitor. Openlayer links a project and inference pipeline, then starts ingesting matching executions. The Agents table shows each chatbot and its number of imported sessions. Enabled AIP Chatbot projects display a Foundry mark in the configured integrations and projects tables. To stop ingesting future traces for a chatbot, open its Agent options menu and select Disable. Existing traces remain available.

​

Step 4: Configure sync

Under Settings, enter the Session-log dataset RID and Stream branch, then click Save session log. Turn on Periodic sync to poll the stream every 15 minutes, or click Sync now to queue a manual sync. Periodic sync requires a dataset RID and the api:streams-read scope.

---

​

Ingest a single session

The session-log stream is the primary sync source. If you have a gateway-originated session’s three identifiers, you can ingest that session directly:

1. Open the chatbot row’s Agent options menu and select Ingest.
2. Enter the Session RID and Session trace ID. The agent RID is already known from the chatbot row.
3. Click Ingest.

This action uses getSessionTrace and only works when you have the IDs from the originating client. Palantir documents that these sessions expire after approximately 24 hours.

---

​

Trace mapping

Openlayer maps Palantir JSON session-log events as follows: Model, token, and cost data appear only when sibling language-model usage events share the same traceId. Chatbot events do not always include these usage events.

---

​

Disconnecting

To disconnect, open Settings → Integrations → Palantir Foundry AIP Agents and click Disconnect. Disconnecting stops session-log syncing and deletes registered agent records. Existing Openlayer traces are preserved.

​

Troubleshooting

Sync fails with a missing dataset RID. Configure Session-log dataset RID under Settings. A Foundry Org Admin must create the export with the Palantir JSON schema. No chatbots appear after discovery. Discovery only finds AIP_AGENTS_AGENT Compass files. Confirm that the token has api:filesystem-read and can see the folders containing your chatbots. You can also register a chatbot by RID or filesystem path. An enabled chatbot has zero traces. Confirm that the session-log export covers the chatbot’s project, uses the Palantir JSON schema, and has started receiving rows. The chatbot must be enabled, and executions remain pending until they include a final_response or execution_error. A single-session ingest is not ready. Confirm that getSessionTrace reports a complete session and that you entered the correct session trace ID. Path A sessions expire after approximately 24 hours.