--- title: "Reportei Integration" description: "Connect AI agents to Reportei, a marketing report and dashboard builder for agencies. Agents generate client reports from templates, pull metrics across connected integrations, and post timeline events that explain a change." url: https://flowrunner.ai/integrations/reportei date_modified: 2026-08-11T09:36:14-07:00 --- # Reportei [Analytics & Data](https://flowrunner.ai/integrations/category/analytics-data) Connect AI agents to Reportei, a marketing report and dashboard builder for agencies. Agents generate client reports from templates, pull metrics across connected integrations, and post timeline events that explain a change. [Verified](https://flowrunner.ai/integrations/verified "What does verified mean?") · 26 actions · API key · available [Reportei website](https://reportei.com/) · [Platform Documentation](https://reportei.com) · Capability data verified 2026-08-05 1. The monthly reporting cycle opens, since Reportei starts no flows itself 2. List Projects returns the client projects due this cycle, paginated 3. List Integrations confirms which data sources are actually connected for each project 4. Get Metric Data returns the period values and the comparison period values per source 5. The agent flags projects where a source is disconnected or a metric returns nothing 6. The account lead receives the numbers and the comparison framing before anything is built 7. Lead approves the period, the comparison, and the narrative before Create Report runs ## What This Integration Enables Reportei is built for agencies, and agencies have a problem that is not really a measurement problem. The metrics are already available somewhere; what costs the month is that thirty clients each need the same assembly done, at the same time, by people whose hours are billable elsewhere. This connector attacks that directly by making the whole object model reachable: projects, which are clients, the integrations connected to each one, the templates that define structure, the metrics available per data source, the client-facing documents and live dashboards created from those templates, timeline events that annotate a project, and webhook subscriptions. FlowRunner agents read company settings and rate limits; list and retrieve projects; list and retrieve connected data sources filtered by project, name, or slug; list and retrieve templates with their section and widget structure; list the queryable metrics for a data source slug and retrieve metric values for a date range with an optional comparison period; list, retrieve, and create reports; list, retrieve, and create live dashboards; list, retrieve, create, update, and delete timeline events; and manage webhook subscriptions. Two boundaries define what an agent can do here. Reports and dashboards can be created from a template for a project and a date range, and they can be listed and retrieved, but there is no update or delete operation for either, so one that exists is one that exists. That is a strong argument for approving the inputs beforehand rather than fixing anything afterwards. And on events, be plain: the connector exposes no FlowRunner triggers. The webhook operations here subscribe a callback URL of your choosing to a Reportei event type, optionally scoped to a project, and List Webhook Event Types tells you what can be subscribed to. That is a Reportei subscription pointing at an external endpoint, not something that starts a flow. Flows begin on a schedule or on your own calendar. Pagination follows a page and per-page model capped at 100 per page, with a meta object carrying page totals. The connector is built and verified against Reportei's official API. ### Without FlowRunner **Reporting week eats the calendar**: The same assembly work repeats for every client at the same time every month **Broken data sources found late**: A disconnected integration is discovered by the client reading a zero **Context lives in the account manager's head**: The reason a metric moved is explained on a call and never written down anywhere ### With FlowRunner **Assembly happens before anyone opens a tab**: Metrics for every project are pulled and checked ahead of the reporting cycle **Source health checked first**: Disconnected integrations are flagged before anything is built on them **Explanations attached to the timeline**: The reason behind a change is recorded as a timeline event on the project it belongs to ## Use Case Scenarios ### Source Health Checked Before the Month Closes Nothing damages an agency's credibility faster than a client-facing document built on a data source that stopped authenticating three weeks ago. On a schedule, the agent calls List Projects to iterate the book and List Integrations per project to enumerate connected sources. For each source it calls List Available Metrics and Get Metric Data over a short recent window. Where a source returns nothing, it flags the project and posts to the account lead in [Slack](https://flowrunner.ai/integrations/slack) with the client, the source, and the last period that returned data. The reconnection happens before the cycle rather than during the client call. ### Numbers Assembled Ahead of the Cycle The cycle opens. The agent calls List Templates to find the template that applies to each client tier, calls Get Metric Data for the period with the comparison period set to the prior equivalent window, and assembles a working summary per project into a sheet in [Google Sheets](https://flowrunner.ai/integrations/google-sheets). The account lead opens the month with the numbers already gathered and a list of the accounts where something actually moved, and spends the time on the ones that need a story rather than on the twenty that do not. ### The Explanation Recorded Where It Belongs A campaign is paused, a budget is shifted, or a tracking change lands. Rather than that context living in a chat thread, the agent calls Create Timeline Event with HTML content describing what changed and when, scoped to the project it affects. Three months later, when a client asks why a metric dipped in that period, the answer is on the timeline instead of in somebody's memory. Where an event needs correcting, Update Timeline Event and Delete Timeline Event let the record be amended, which is the difference between an annotation and a permanent artefact. ## Human-in-Loop Highlight Create Report is the moment the agency stops talking to itself. A document built from a template for a project and a date range goes to the client, and this connector offers no operation to update or delete one once it exists, which means the editorial decisions get made before the call, not after. The decision that matters most is not the metric, it is the comparison. Get Metric Data takes a date range and an optional comparison period, and the choice of that comparison is what turns a number into a claim. Set against last month, a campaign looks like a recovery. Set against the same month last year, the identical number looks like a decline. Both are true, and picking one is an act of account management that an agent has no standing to perform on a client's behalf. So the agent prepares both and hands the choice over. It calls Get Metric Data for the period and for each candidate comparison, and posts to the account lead: "Ready to build for \[client\] from template \[name\], period \[range\]. Headline metrics: \[list with values\]. Against the prior period: \[deltas\]. Against the same period last year: \[deltas\]. Sources returning no data this period: \[list\]. Proposed timeline events to include: \[drafts\]. Confirm the comparison period and approve, or amend?" The lead picks the framing, and only then does the agent call Create Report and Create Timeline Event. Reading metrics, checking source health, and flagging the accounts that moved all run unattended across the whole book, because none of it reaches a client. Deciding what a number means, in something that goes outside the company and cannot be edited afterwards, is [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) work. Agent processes routinely Detects exception requiring judgment Clear match Continues automatically Ambiguous Routes to human via preferred channel Human decides Agent resumes with decision ## Agent Capabilities 26 actions ### Company 1 - **Get Company Settings** Retrieves the company profile and the API rate limits. Used to pace a sweep across a large book of clients. ### Projects 2 - **List Projects** Lists projects, which are clients, with pagination and sorting. Used to iterate the whole book on a cycle. - **Get Project** Retrieves a single project by id. ### Integrations 2 - **List Integrations** Lists the connected data sources, filterable by project, name, or slug. This is the source health check: a project missing an expected integration is about to produce a zero in front of a client. - **Get Integration** Retrieves a single integration by id. ### Templates 2 - **List Templates** Lists the reusable templates. Used to select the right structure per client tier. - **Get Template** Retrieves a single template with its section and widget structure. Used to know which metrics will actually be needed before pulling them. ### Metrics 2 - **List Available Metrics** Lists the queryable metrics for a data source slug. Used to confirm what can be measured before promising it. - **Get Metric Data** Retrieves metric values for a date range, with an optional comparison period. The comparison period is the editorial choice in client-facing output, which is why it belongs in an approval rather than in a default. ### Reports 3 - **List Reports** Lists reports with pagination and sorting. - **Get Report** Retrieves a single report by id. - **Create Report** Creates a report from a template for a project and date range. There is no update or delete operation for reports, so the inputs are the decision. ### Dashboards 3 - **List Dashboards** Lists live dashboards with pagination and sorting. - **Get Dashboard** Retrieves a single dashboard by id. - **Create Dashboard** Creates a live dashboard from a template for a project and date range. Like reports, dashboards have no update or delete operation here. ### Timeline Events 5 - **List Timeline Events** Lists timeline events, which are annotations on a project or a report. - **Get Timeline Event** Retrieves a single timeline event by id. - **Create Timeline Event** Adds a timeline event with HTML content to a project or report. Used to record why a metric moved at the moment it moves, rather than reconstructing it later. - **Update Timeline Event** Updates an existing timeline event. This is what makes an annotation correctable rather than permanent. - **Delete Timeline Event** Deletes a timeline event. ### Webhooks 6 - **List Webhooks** Lists the configured webhook subscriptions and their destinations. - **List Webhook Event Types** Lists the event types available for subscription. Used to discover what Reportei can notify an external endpoint about. - **Get Webhook** Retrieves a single webhook by id. - **Create Webhook** Subscribes a callback URL to an event, optionally scoped to a project. This creates a Reportei subscription pointing at an endpoint of your choosing; it does not create a FlowRunner trigger. - **Update Webhook** Updates a webhook's URL, event, or project filter. - **Delete Webhook** Deletes a webhook subscription. ## Frequently Asked Questions ### What can FlowRunner do with Reportei? FlowRunner agents can run Get Company Settings, List Projects, and Get Project in Reportei, plus 23 more actions. ### Does connecting Reportei to FlowRunner require OAuth? No. Reportei connects to FlowRunner with an API key, no OAuth flow required. ### Can Reportei trigger a FlowRunner workflow automatically? Reportei doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger. **Work at Reportei?** This integration exposes Reportei to AI agents on every FlowRunner plan, including through MCP, at no cost to you. [See what FlowRunner offers integration partners](https://flowrunner.ai/integrations/partners), including how to keep this page current. --- Markdown version of https://flowrunner.ai/integrations/reportei. Site index: https://flowrunner.ai/llms.txt