RankYak
SEOPull AI-written SEO articles out of RankYak and push them into your own publishing pipeline. Agents route drafts to an editor before anything goes live under your byline.
What This Integration Enables
RankYak's entire API is three read and write endpoints, and none of them commissions or edits an article. That constraint is the most useful thing to know about this integration. RankYak plans and writes on its own content calendar whether or not anyone is watching, and the only lever available from a flow is what happens between its calendar and your domain. So this connector is deliberately narrow: it collects finished work, moves it into your pipeline, and reports back where it landed. List Articles returns the planned, written and published backlog with scheduling fields that distinguish them, Get Article retrieves one piece with the full body in both HTML and Markdown, and Store Article External URL records the public URL your CMS produced so RankYak can associate performance with the article it wrote.
Everything the connector returns is publication-ready and none of it has been read by anyone on your team. An article carries a title, SEO meta fields, an excerpt, a header image and a body that will render on your domain under your name. Two fields decide whether the drop is clean: the archived timestamp, which marks articles that remain live but stop getting refreshed and are usually not worth re-rendering, and the difference between the scheduled slot and the real publication time. Treating that handoff as a pipeline with a person in it rather than a pipe from one system to another is the whole point of this integration, and it is why human-in-the-loop belongs between RankYak's calendar and your CMS rather than after publication.
Without FlowRunner
With FlowRunner
Use Case Scenarios
Calendar to CMS Draft, Every Time
On Article Published fires. The agent reads the payload, checks the archived timestamp, and creates the piece in WordPress as a draft using the HTML body, with Exclude H1 turned on because the theme renders the title itself and would otherwise print it twice. It attaches the header image, maps the SEO title and description into the site's meta fields, and posts the draft link to the editorial channel in Slack. Nothing is live. The editor opens a formatted draft in the tool they already use rather than a block of text pasted from another dashboard.
Backfilling an Existing Backlog Into an Editorial View
A team inherits an account with months of RankYak output already written. The agent pages through List Articles at 100 per page, filters out archived pieces, and creates a row per article in Notion with the title, excerpt, scheduled slot and current publication state. The result is an editorial calendar the team can triage: what is planned, what is written and waiting, what already went live somewhere. The backlog becomes a queue with owners instead of an archive nobody has read.
Closing the Attribution Loop After a Human Publishes
An editor approves a draft and publishes it. The CMS webhook fires, the agent picks up the permalink, and calls Store Article External URL for that article ID. RankYak now knows which live page corresponds to which article it wrote, which is the only way it can attribute performance to the piece. Where the editor spiked the draft instead, the agent records nothing and closes the item as declined, so RankYak's ledger reflects what actually happened on the site rather than what it assumed.
Human-in-Loop Highlight
By the time this flow sees an article, RankYak already considers it published. The remaining question is not whether the text exists, it is whether it becomes yours: your domain, your byline, your topical footprint. Google's helpful content guidance and plain brand safety both point at the same answer, which is that a person should read it first. So the agent never creates a live post, only a draft, and the message to the editor says exactly what state things are in: "New RankYak article ready: [title]. Target keyword [keyword]. Rendered as a draft at [CMS link], not published. Publishing puts this on [domain] under the site byline, and Store Article External URL will only run once you do." The editor publishes, rewrites or spikes it. The agent handles the fetch, the formatting and the attribution callback, and stops at the one decision that puts a machine-written page on a domain a person is accountable for.
Agent Capabilities
3 actionsArticles
3- List Articles Returns the articles RankYak has planned, written or published for the account, each with title, SEO meta fields, excerpt, header image and the full body in HTML and Markdown, paginated up to 100 per page. Used to sync a whole backlog into a CMS or an editorial database, with the scheduling fields separating planned pieces from live ones.
- Get Article Retrieves a single article by ID with both body formats, the SEO title and description, the header image URL and every timestamp RankYak tracks. Used when a flow already has an article ID and needs the current version, and the place to check the archived timestamp before re-rendering anything.
- Store Article External URL Records the public URL where an article was actually published on your own site. Used immediately after a human-approved publish, since RankYak cannot connect an article to its live page or attribute any performance to it without this call.
Triggers
1 triggersEvent Triggers
1- On Article Published Fires when RankYak publishes a new article on its content calendar, delivering the finished piece with its title, SEO meta fields and body in both HTML and Markdown. RankYak's webhooks are configured in its dashboard with no API to manage them, so this trigger polls the article list and reports articles whose published timestamp it has not seen before, at a polling interval of 30 seconds or longer. Used to start the draft-and-review flow the moment new content exists, rather than on a schedule that may run hours after the fact.
Frequently Asked Questions
What can FlowRunner do with RankYak?
FlowRunner agents can run List Articles, Get Article, and Store Article External URL in RankYak.
Does connecting RankYak to FlowRunner require OAuth?
No. RankYak connects to FlowRunner with an API key, no OAuth flow required.
Can RankYak trigger a FlowRunner workflow automatically?
Yes. RankYak supports 1 trigger that can start a FlowRunner workflow automatically.
Start building with RankYak
$100 in credits. No card required. Connect in minutes.