Skip to main content
The KIME API lets your own systems read your KIME data, with no person or AI assistant in the loop. Use it when something should run on its own: on a schedule, inside your own product, or synced with another tool. See API access for keys, endpoints and request details.
The API is read-only. It can list workspaces, brands, prompts, categories and tags, and read key metrics, source domains and source URLs. It cannot create, edit or delete anything. To make changes from outside the app, such as adding prompts or updating tasks, use the MCP Server. If you need write access through the API itself, contact team@kime.ai.

​
What the API gives you

All analytics endpoints can be filtered by date range, answer engine, brand, prompt, category and tags.

​
Dashboards and white-label reporting

Pull visibility, placement, sentiment and share of voice into your own BI tool or internal dashboard instead of sending people to log in to KIME. An agency can loop through its client workspaces and build a branded view for each client. For a no-code option, use the Data Studio connector.

​
Scheduled reports and digests

Call the metrics endpoints on a schedule (for example a weekly job) and send the results where your team reads them: a Slack or Teams message, an email, a PDF or a spreadsheet. A job can rank every tracked brand or workspace and post a summary automatically.

​
Threshold alerts

Poll the metrics on a schedule, compare with the previous period, and send a notification when something crosses a threshold you set. For example:
  • Your sentiment score drops by more than a set amount.
  • A competitor’s share of voice overtakes yours.
  • Your visibility on a category of prompts falls.

​
Source and outreach research

Use source domains and source URLs to build a list of the sites and pages AI cites most in your category, and which brands they mention. Feed it into your PR or content planning tools to decide where to earn mentions.

​
Data warehouse

Load KIME metrics into your data warehouse (for example Snowflake, BigQuery or Redshift) on a schedule and join them with your CRM, product or revenue data. This lets you answer questions like whether a rise in AI share of voice is followed by more sign-ups or pipeline.

​
Embedding KIME data in your own product

If your product serves a marketing, SEO or brand audience, the API can power an “AI visibility” feature inside your app. Your product handles the UI and authentication, and calls the API server-side.

​
Notes for building on the API

  • Authentication, pagination and dates: see API access.
  • Sentiment scale: 0 to 100, where higher is more positive.
  • Engine filtering: filter by answer engine to build an engine-by-engine breakdown.

​
When to use MCP instead

Use the MCP Server when a person wants to ask questions in plain language, or when you need to make changes, such as adding prompts in bulk, archiving prompts, adding competitors, or creating and updating tasks. See MCP use cases.