Reference guides for the analytics apps an AI agent can work with: every endpoint, the permission each one needs, and how to give agents safe, governed access.
The Amplitude API is how an app or AI agent works with a product analytics project: sending events, querying analytics such as segmentation, funnels, and retention, exporting the raw event stream, and managing cohorts and the event taxonomy.
The Fullstory API is how an app or AI agent works with a Fullstory account: listing a user's sessions, reading the events captured in a session, sending users and custom events back in, and generating AI summaries of what a person did.
The Google Analytics API is how an app or AI agent works with a Google Analytics 4 property: running a report over event data, pulling realtime activity, reading account and property configuration, and creating or editing data streams and custom fields.
The Heap API is how an app or AI agent sends data into a Heap environment: recording a custom server-side event, tying an anonymous visitor to a known identity, attaching properties to a user or an account, and submitting users for deletion.
The Matomo API is how an app or AI agent works with a Matomo analytics install: reading visit and page reports, pulling real-time and raw per-visit detail, managing goals, and adding or configuring tracked sites.
The Metabase API is how an app or AI agent works with a Metabase instance: running a saved question, running an ad-hoc query against a connected database, building dashboards and collections, and managing users and permissions groups.
The Mixpanel API is how an app or AI agent works with a Mixpanel project: recording events, updating the people and group profiles those events belong to, and running segmentation, funnel, and retention reports over what has been collected.
The Pendo API is how an app or AI agent works with a Pendo subscription: querying product usage across visitors and accounts, reading and writing visitor and account metadata, and listing the features, pages, and guides being tracked.
The Plausible API is how an app or AI agent works with a Plausible account: querying a site's visitors and conversions, creating or deleting a site, managing its goals and shared links, or recording a pageview from a server.
The PostHog API is how an app or AI agent works with a PostHog project: sending events in, running HogQL queries to read them back, reading and updating the people behind the analytics, and managing feature flags, insights, and experiments.
The Power BI API is how an app or AI agent works with a Power BI tenant: listing and refreshing datasets, exporting a report to a file, creating a workspace, or uploading a published file.
The Segment API is how an app or AI agent works with a Segment workspace: sending a customer event into a source, creating a source or destination, editing a tracking plan, or deploying a function.
The Tableau API is how an app or AI agent works with a Tableau site: listing workbooks and views, pulling the data or an image behind a view, publishing and refreshing data sources, and managing projects, users, and groups.
An analytics API is how an app or AI agent reads product and web data, or sends events into it. The platforms here, including Amplitude, Mixpanel, PostHog, and Google Analytics, hold records of what people do in a product, and their APIs let an agent query those numbers, export a report, or push new events in as they happen.
Most analytics access is read access to aggregated figures, but some keys can also send events or change dashboards, and a few reach raw event records that name individual people. Access runs on an access key or token, and the permission attached to it sets whether an agent can only read, can also write events, or can manage the account. A governed layer means an agent that only needs to read a funnel never holds a key that can also alter tracking.
Bollard AI sits between a team's AI agents and the apps it runs on. Grant each agent exactly the access it needs, read or write, app by app, and every call is checked and logged.