docs(TSP-1335): document evaluation version picker and MCP tool

build/agents/build-your-agent/evals.mdx

@@ -1,18 +1,18 @@
 ---
 title: 'Evals'
 sidebarTitle: 'Evals'
 description: 'Test and evaluate your AI Agents with scenario-based evaluations and reusable Checks'
 ---
 
 <Info>
 **Rollout Status**: Evals is rolling out progressively, starting with Enterprise customers. If you don't see this feature in your account yet, reach out to your account manager to discuss access.
 </Info>
 
 The Evals section is your command center for testing and evaluating AI Agent performance. Located in the **Evaluate** tab (next to the Build and Use tabs) in the Agent builder, Evals lets you create test sets, define reusable Checks, run automated evaluations, and monitor live Agent quality — all without manual testing. Evals apply to both individual Agents and Workforces, including the sub-agents and tools inside a Workforce.
 
 ![Evaluate tab showing the Evals sidebar (Test, Runs, Checks, Publish, Monitor) and a Monitor dashboard with overall score, total runs, and Checks breakdown](/images/agent/agent-evals.png)
 
 ## What you can do with Evals
 
 <CardGroup cols={3}>
   <Card title="Run tests" icon="flask-vial">
@@ -28,11 +28,11 @@
 
 ---
 
 ## Evals sections
 
 The Evals area has five sections, shown in the left sidebar of the Evaluate tab:
 
 - **Test** — Create and manage test sets. Each test set holds scenarios that simulate users; running a scenario produces a conversation with your Agent that gets scored by attached Checks.
 - **Runs** — Past evaluation run results. Browse average scores, tasks evaluated, progress status, cost (Credits and Actions), and creation date for every run.
 - **Checks** — The reusable set of evaluation criteria. Create a Check once, then attach it to scenarios, to Monitor dashboards, or to one-off evaluations of completed tasks.
 - **Publish** — Choose which test sets must pass before your Agent can be published. Set a minimum pass rate and optionally block publishing on failure.
@@ -216,8 +216,23 @@
 You can select specific scenarios within a test set to run a subset at once, or run all scenarios in the test set together. Note that you cannot bulk-select and run multiple test sets at the same time.
 
 1. Enter a name for the run (e.g., "Scenario run - Jan 14, 12:14 PM"). A default name with timestamp is provided.
-2. Checks already attached to the scenarios are always included. To add Checks from the Checks tab, select the ones you want under **Additional global checks**.
-3. Click **Run** to begin. The simulator generates conversations with your Agent based on your scenario prompts and the selected Checks score each conversation.
+2. Select which version to evaluate from the **Version** dropdown. It defaults to the current active version. Select any historical or draft version to test against instead. This applies to both individual agents and workforces.
+3. Checks already attached to the scenarios are always included. To add Checks from the Checks tab, select the ones you want under **Additional global checks**.
+4. Click **Run** to begin. The simulator generates conversations with your Agent based on your scenario prompts and the selected Checks score each conversation.
+
+### Running evaluations via MCP
+
+You can trigger evaluation runs programmatically using the `relevance_run_evaluation` MCP tool available through the [Relevance AI MCP server](/integrations/mcp/programmatic-gtm/mcp-server). This mirrors the version picker available in the run modals.
+
+| Parameter | Required | Description |
+|-----------|----------|-------------|
+| `agent_id` | Yes | The ID of the agent or workforce to evaluate |
+| `test_set_id` | Yes | The ID of the test set to run |
+| `version_id` | No | A specific agent version ID to evaluate against. Omit to use the current active version |
+
+<Note>
+Workforce evaluations now correctly run against the version specified — either via the Version dropdown in the run modal or the `version_id` parameter via MCP. Previously, workforce evaluations always used the active published version regardless of the selection.
+</Note>
 
 ---
 
@@ -233,7 +248,7 @@
 |--------|-------------|
 | **Average Score** | Overall pass rate across all scenarios and Checks |
 | **Tasks** | How many Agent tasks were evaluated |
-| **Agent Version** | The version of the Agent that was tested |
+| **Agent Version** | The version of the agent or workforce that was evaluated, as selected in the run modal |
 
 ### Scenario results
 
@@ -290,7 +305,7 @@
 
 You also get:
 
 - **Overall score timeseries** to spot regressions or improvements over time.
 - **Per-Check charts** so you can see which criteria are slipping.
 - **Version markers** that line up score changes with Agent publishes.
 - **A list of evaluation runs** with score, name, and a drill-in to the full conversation.
@@ -389,7 +404,7 @@
   </Accordion>
 
   <Accordion title="Can I rerun a previous evaluation?">
-    Yes, you can run the same scenarios again at any time. Each run is saved in your Runs history, allowing you to compare results across different Agent versions.
+    Yes, you can run the same scenarios again at any time. Each run is saved in your Runs history. You can also target a specific historical or draft version using the **Version** dropdown in the run modal, letting you compare results across different agent versions directly.
   </Accordion>
 
   <Accordion title="Where do my Checks live?">

integrations/mcp/programmatic-gtm/mcp-server.mdx

@@ -39,7 +39,7 @@
   </Accordion>
 
   <Accordion title="ChatGPT" icon="comment">
     ChatGPT supports MCP servers through Developer Mode, available on Pro, Team, Enterprise, and Edu plans.
 
     1. Open ChatGPT **Settings**
     2. Go to **Connectors** → **Advanced** → **Developer Mode**
@@ -242,6 +242,19 @@
 
 ---
 
+## Available tools
+
+The Relevance AI MCP server exposes the following tools to connected AI clients:
+
+| Tool | Description |
+|------|-------------|
+| `relevance_trigger_agent` | Triggers an agent synchronously and waits up to 120 seconds for the result. |
+| `relevance_trigger_agent_async` | Triggers an agent and returns immediately with a `conversation_id` for polling. Use for long-running agents. |
+| `relevance_poll_agent_result` | Polls for the result of an async agent run using the `conversation_id`. |
+| `relevance_run_evaluation` | Runs an evaluation test set against an agent or workforce. Accepts `agent_id` (required), `test_set_id` (required), and an optional `version_id` to evaluate a specific historical or draft version instead of the current active version. |
+
+---
+
 ## Handling long-running agent executions
 
 When triggering agents via MCP, you have two execution modes available depending on how long your agent takes to complete.