invarium_get_agent
Get detailed information about a specific agent. Returns tools, constraints, AQS score with trend, recent test runs, and a dashboard link.
When to Use
Call invarium_get_agent when you need to inspect a specific agent’s current state. Common scenarios:
- After syncing test results, to see the updated AQS score and trend
- Before generating new tests, to review the agent’s tools and constraints
- When diagnosing test failures, to check the agent’s configuration
- To get a quick link to the agent’s dashboard page
See Agent Quality Score for details on how AQS is calculated and how to interpret trends.
Parameters
| Name | Type | Required | Default | Description |
|---|---|---|---|---|
agent_name | string | Yes | — | Name of the agent to look up. Must match the name used when the blueprint was uploaded. |
invarium_get_agentGet detailed information about a specific agent. Returns tools, constraints, AQS score, recent test runs, and a dashboard link.
Parameters
| Name | Type | Required | Description |
|---|---|---|---|
agent_name | string | required | Name of the agent to look up. |
Returns
Formatted agent details including status, AQS score with trend, tools, constraints, recent test runs, and dashboard link.
Example
Agent: customer-support-agent
Status: ACTIVE
Description: Handles customer inquiries via chat
AQS Score: 78/100 (↑ +5)
Tools (3):
- search_tickets: Search support tickets by keyword
- create_ticket: Create a new support ticket
- escalate_to_human: Transfer conversation to a human agent
Constraints (2):
- Never share internal ticket IDs with customers
- Escalate billing disputes to a human agent
Recent test runs: 3 completed, 1 running
Last run: 2 hours ago (passed: 8, failed: 2)
View on dashboard: https://app.invarium.dev/agent/ag_abc123Response
The tool returns a formatted text block with the agent’s full details:
Agent: customer-support-agent
Status: ACTIVE
Description: Handles customer inquiries via chat
AQS Score: 78/100 (↑ +5)
Tools (3):
- search_tickets: Search support tickets by keyword
- create_ticket: Create a new support ticket
- escalate_to_human: Transfer conversation to a human agent
Constraints (2):
- Never share internal ticket IDs with customers
- Escalate billing disputes to a human agent
Recent test runs: 3 completed, 1 running
Last run: 2 hours ago (passed: 8, failed: 2)
View on dashboard: https://app.invarium.dev/agent/ag_abc123Response Fields
| Field | Description |
|---|---|
| Agent | The agent’s registered name. |
| Status | Current status: ACTIVE, INACTIVE, or UNKNOWN. |
| Description | The agent’s description from its blueprint. |
| AQS Score | Agent Quality Score (0—100) with trend indicator. See below. |
| Tools | List of registered tools with their descriptions. |
| Constraints | List of behavioral constraints and guardrails. |
| Recent test runs | Summary of recent test runs by status (completed, running, etc.). |
| Last run | When the most recent test run happened and its pass/fail counts. |
| Dashboard link | Direct URL to the agent’s page on the Invarium dashboard. |
AQS Score and Trend
The AQS (Agent Quality Score) reflects the agent’s overall behavioral quality based on test results. The trend indicator shows the change since the previous measurement:
| Display | Meaning |
|---|---|
78/100 | Current score, no trend data available yet. |
78/100 (↑ +5) | Score improved by 5 points since the last measurement. |
78/100 (↓ -3) | Score decreased by 3 points since the last measurement. |
-- | No test runs have been completed yet. AQS is not available until the first test cycle. |
Examples
Basic Usage
Look up a specific agent by name:
"Show me details for my customer-support-agent."After Syncing Results
Check the updated AQS score after syncing test results:
"Sync these results for customer-support-agent, then show me its updated AQS score."Reviewing Before Test Generation
Check the agent’s current tools and constraints before generating new tests:
"Show me the tools and constraints for my travel-booking-agent before I generate new tests."Error Responses
| Error | Cause | Fix |
|---|---|---|
Authentication failed: invalid API key | Invalid or missing API key. | Verify your INVARIUM_API_KEY. Run invarium_connect first. |
Agent '{name}' not found. Upload a blueprint first with invarium_upload_blueprint. | No agent with this name exists in your workspace. | Check the agent name. Use invarium_list_agents to see all registered agents. |
Failed to get agent: ... | Backend API error. | Check the error details and try again. |
See Error Codes for the full error reference.