> ## Documentation Index > Fetch the complete documentation index at: https://docs.ringg.ai/llms.txt > Use this file to discover all available pages before exploring further. > ## Agent Instructions > To search or read these docs, connect the Ringg docs MCP server at https://docs.ringg.ai/mcp (no key needed; tools search_ringg_ai and query_docs_filesystem_ringg_ai) instead of browsing pages one by one. Setup for each client: https://docs.ringg.ai/resources/mcp-server/docs-mcp-server/mcp-for-ai-assistants.md > The dashboard labels agents "Assistants"; these docs say "agent". Both mean the same object, identified by agent_id. > API base URL: https://prod-api.ringg.ai/ca/api/v0. Send the workspace API key in the X-API-KEY header, from server-side code only. > Placing a call dials a real phone number and spends credits. Confirm the number with the user before calling. > To act on a Ringg workspace (place calls, edit agents, read transcripts), use the separate Ringg workspace MCP server at https://mcp.ringg.ai/mcp: https://docs.ringg.ai/resources/mcp-server/workspace-mcp-server/overview.md > Get goal rate, latency and cost for your agents, a side-by-side comparison of agent versions, and call, minute and cost totals for each agent in scope. # Get Agent Performance export const RelatedTopics = ({items}) =>
Related topics
{items.map(([title, href]) => {title} )}
; Returns the **Performance** cards and the agents table in [Analytics](/documentation/monitor/analytics). Pass two or more IDs in `agent_version_id` to compare versions in `version_comparison`. ## Notes * `summary` covers every version of the selected agents; `agent_version_id` changes only `version_comparison`. * `goal_rate` uses each call's own version goal (see [Where the goal comes from](/documentation/monitor/analytics#where-the-goal-comes-from)). It is `null` when no call in scope has a goal value. * `total_cost`, `cost` and `total_minutes` count completed calls only, in the workspace `currency`. * `client_analysis_key`, `client_analysis_value` and `client_analysis_type` narrow the result to calls with one [Advanced Analysis](/documentation/build-agents/customize-agent-behaviour/advanced-analysis) value, for example `disposition` equal to `interested`. * `assistants_in_action` lists agents with at least one call in scope, sorted by `call_count`. ## Errors * `400`: a filter is invalid, such as an unknown `channel`, a malformed ID, or `sub_workspace_id` sent from a workspace that isn't a primary workspace. * `401`: `X-API-KEY` is missing or invalid. ## OpenAPI ````yaml get /analytics/v4/agent-performance openapi: 3.0.0 info: title: Ringg AI API Documentation description: >- The Ringg AI REST API. Base URL https://prod-api.ringg.ai/ca/api/v0; send your workspace API key in the X-API-KEY header from server-side code only. Requests and responses are JSON. version: 2.0.0 contact: name: Ringg AI support email: admin@ringg.ai url: https://docs.ringg.ai servers: - url: https://prod-api.ringg.ai/ca/api/v0 security: - ApiKeyAuth: [] tags: - name: workspace description: Endpoints for managing your workspace. - name: agent description: Endpoints for managing agents. - name: calling description: Endpoints for making and managing calls. - name: campaign description: Endpoints for managing campaigns. - name: analytics description: >- Analytics v4 endpoints for call, agent, cohort and chat performance metrics. - name: termination description: Endpoints for terminating active calls using different methods. - name: knowledgebase description: Endpoints for managing knowledge bases. paths: /analytics/v4/agent-performance: get: tags: - analytics summary: Get Agent Performance description: >- Returns goal rate, latency and cost for the selected agents, a side-by-side comparison of agent versions, and per-assistant call, minute and cost totals. operationId: getAgentPerformance parameters: - name: start_date in: query description: >- Start of the reporting window. Accepts `YYYY-MM-DD` or ISO-8601. Omit both dates for all time. required: false schema: type: string example: '2026-10-01' - name: end_date in: query description: >- End of the reporting window (inclusive). Accepts `YYYY-MM-DD` or ISO-8601. required: false schema: type: string example: '2026-10-15' - name: agent_id in: query description: Filter by one or more agent IDs (comma-separated). required: false schema: type: string example: 00000000-0000-4000-8000-0000000000a1 - name: agent_version_id in: query description: Filter by one or more agent version IDs (comma-separated). required: false schema: type: string example: 00000000-0000-4000-8000-0000000000d1 - name: bulk_list_id in: query description: >- Filter by one or more campaign (bulk list) IDs (comma-separated). Callback calls inherit the campaign of their parent call. required: false schema: type: string example: 00000000-0000-4000-8000-0000000000c1 - name: voicemail in: query description: >- `true` returns only voicemail calls, `false` excludes them. Omit to include all calls. required: false schema: type: boolean example: false - name: channel in: query description: >- Filter by call channel (comma-separated): `inbound`, `outbound`, `webcall`. Unknown values return `400`. required: false schema: type: string example: outbound - name: include_callback in: query description: >- Include callback-feature calls (`call_category = CALLBACK`) in the metrics. Defaults to `false`. required: false schema: type: boolean default: false example: false - name: client_analysis_key in: query description: >- Only include calls whose post-call analysis field has this key, for example `disposition`. Use with `client_analysis_value`. required: false schema: type: string example: disposition - name: client_analysis_value in: query description: Value that `client_analysis_key` must equal. required: false schema: type: string example: interested - name: client_analysis_type in: query description: How to compare `client_analysis_value`. required: false schema: type: string enum: - string - boolean - number default: string - name: sub_workspace_id in: query description: >- Primary workspaces only: limit the combined analytics to these Ringg Spaces. Repeat the parameter for multiple IDs. The primary workspace's own ID is accepted; any other workspace sending this parameter gets `400`. required: false schema: type: array items: type: string format: uuid example: - 00000000-0000-4000-8000-000000000101 style: form explode: true responses: '200': description: Agent performance retrieved successfully. content: application/json: schema: type: object properties: summary: type: object properties: goal_rate: type: number description: >- Share of scored calls where the agent version's goal key is `true` (0-1). `null` when nothing is scored. nullable: true avg_latency_ms: type: number description: Average response latency in milliseconds. nullable: true total_cost: type: number description: Credits spent on completed calls. cost_per_outcome: type: number description: '`total_cost` divided by calls that achieved the goal.' nullable: true latency: type: object properties: p50_ms: type: number description: Median latency in milliseconds. nullable: true p95_ms: type: number description: 95th-percentile latency in milliseconds. nullable: true scored: type: integer description: Calls with a latency measurement. total: type: integer description: Calls in scope. currency: type: string description: >- Your workspace's billing currency, used by the cost fields. version_comparison: type: array description: >- Per-version metrics. Pass two or more IDs in `agent_version_id` to compare versions. items: type: object properties: agent_version_id: type: string description: Agent version ID. total: type: integer description: Calls on this version. connected: type: integer description: Completed calls. connection_rate: type: number description: '`connected / total` (0-1).' nullable: true aht_seconds: type: number description: Average handle time in seconds. nullable: true goal_rate: type: number description: Goal rate (0-1). nullable: true positive_sentiment_share: type: number description: Share of scored calls with positive sentiment (0-1). nullable: true avg_latency_ms: type: number description: Average latency in milliseconds. nullable: true assistants_in_action: type: array description: >- Agents with at least one call in scope, sorted by `call_count` descending. items: type: object properties: agent_id: type: string description: Agent ID. agent_name: type: string description: Agent name. call_count: type: integer description: Calls in scope, across every status. total_connected: type: integer description: Completed calls. total_minutes: type: number description: Talk minutes on completed calls. cost: type: number description: Credits spent on completed calls, in `currency`. example: summary: goal_rate: 0.625 avg_latency_ms: 912.4 total_cost: 42.75 cost_per_outcome: 0.57 latency: p50_ms: 883.4 p95_ms: 1074.7 scored: 102 total: 120 currency: INR version_comparison: - agent_version_id: 00000000-0000-4000-8000-0000000000d1 total: 120 connected: 84 connection_rate: 0.7 aht_seconds: 41.2 goal_rate: 0.625 positive_sentiment_share: 0.72 avg_latency_ms: 912.4 assistants_in_action: - agent_id: 00000000-0000-4000-8000-0000000000a1 agent_name: Ananya call_count: 120 total_connected: 84 total_minutes: 310.5 cost: 42.75 '400': description: >- Bad Request - invalid filter value (for example an unknown `channel` or malformed ID). content: application/json: example: message: 'invalid channel(s): sms' '401': description: Unauthorized - Invalid or missing authentication. '500': description: Internal Server Error. content: application/json: example: message: Failed to fetch agent performance! components: securitySchemes: ApiKeyAuth: type: apiKey in: header name: X-API-KEY ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.