YourTextGuru
SEOPull semantic SEO guides, content scores, and AI writing from YourTextGuru through its V1 API. Agents score a draft against the target query before anyone publishes it.
What This Integration Enables
Most AI content tools produce text. YourTextGuru is unusual because it also grades it, and that grading surface is the reason to wire it in. A guide is the reference object everything else works against: generated from a search query in a given locale, it holds the weighted terms, entities and structure a page needs to compete. Check Content Optimization then scores an arbitrary piece of text against that guide and returns two ratings rather than one, semantic optimization and over-optimization, with each term marked as healthy, under-used or over-used. That second rating is the part worth paying attention to. A tool that only told you to add more terms would be an instruction to stuff the page, and this one is built not to.
The generation surface exists too, and it is deliberately guide-aware. Generate Optimized Text writes steered by a specific guide so the output already carries the vocabulary the query needs, Generate Full Article runs the automatic writing mode against a guide for a complete first draft, Generate Outline produces a heading structure that reflects the subtopics the query demands, and Rephrase Text rewrites a passage while keeping the guide's vocabulary intact. Around all of that sits the cluster layer: groups collect related guides, Build Cocon computes the internal-linking silo that ties them together, and Get Cocon returns which pages should link to which. Note the asymmetry the API creates and this connector inherits. The same guide can both write the page and score it, which is a useful loop and a circular one, and it is why the human-in-the-loop moment on this integration sits on the score rather than on the draft.
Without FlowRunner
With FlowRunner
Use Case Scenarios
Every Draft Scored Before It Reaches the Editor
A writer marks a draft ready in WordPress. The agent reads the body and the target query, finds the existing guide for that query with List Guides filtered to API-created guides, or creates one with Create Guide and polls Get Guide until it reports ready. It runs Check Content Optimization and attaches the result to the draft: the semantic score, the over-optimization score, the terms marked under-used and the terms marked over-used. It also runs Suggest Entities, since entity coverage is a separate signal from keyword coverage and a page can score well on one while missing the people, places and brands that make it read as authoritative. The editor opens a draft with a diagnosis on it, not a verdict.
Planning a Cluster and Its Internal Links in One Pass
A new topic area is approved. The agent calls Create Group, then Create Guide for each query in the cluster with the group attached, polling until they build. It runs Get Guide Questions and Get Guide Related Searches per guide to collect the FAQ material and neighboring intents, and Get Guide SERP to record which competing URLs each guide was built from. Once the cluster is complete it calls Build Cocon and reads the result with Get Cocon, which returns the internal-linking structure across the whole silo. The outline set and the link plan land in Notion as one brief, so the writers know both what to cover and what each piece should link to before anyone starts typing.
Refreshing Pages That Have Drifted From Their Query
A quarterly refresh targets pages that have not been touched in a year. The agent pulls each page's text, scores it with Check Content Optimization against a freshly generated guide, and sorts the results by how far the current page sits from what the live results now demand. For the pages that need work, it runs Rephrase Text on the weakest sections and Generate Questions Section to add an FAQ block built from real search questions. Every rewritten section is re-scored before it is offered, and the whole set goes to the content lead in Slack as a ranked worklist rather than a batch of automatic edits.
Human-in-Loop Highlight
The gate here is on a passing score, which is a stranger place to put it than a delete or a publish, and it is the right one for this connector. A draft can be pushed over an optimization threshold by adding the guide's weighted terms until the number goes green, and the resulting page can be simultaneously well optimized and not worth reading. The over-optimization rating catches part of that, which is why the agent always reports both numbers together. The rest is a judgment no API returns. There is also a circularity worth naming out loud: when Generate Optimized Text wrote the draft against the same guide Check Content Optimization is now grading it with, a high score is partly a description of the tool agreeing with itself. So the agent puts all of it in front of the editor: "[Title] scored [semantic rating] with over-optimization at [rating]. Terms under-used: [list]. Terms flagged over-used: [list]. Entities missing: [list]. This draft was generated with Generate Optimized Text against this same guide. Publish, send back for a rewrite, or drop the topic?" The agent measured the page. A person decides whether the page should exist.
Agent Capabilities
26 actionsGuides
6- Create Guide Starts generation of a semantic optimization guide for a search query in a given locale, optionally attached to a group. Used as the first step of any scoring or writing flow, since the guide is the reference every other action works against.
- Get Guide Retrieves a generated guide with its optimization scores and the weighted term list a page should cover. Used as the polling step after Create Guide, because guides take a few minutes to build and a fresh one may still report in progress.
- List Guides Returns the account's guides with filters for type, locale, group and generation status. Used to poll only the guides still building, or to restrict a flow to API-created guides so manual dashboard work stays out of an automated run.
- Get Guide Questions Returns the "People Also Ask" questions Google surfaces for the guide's query. Used to build FAQ blocks and subheadings from questions searchers actually ask rather than ones a writer imagined.
- Get Guide Related Searches Returns the related searches shown at the foot of the results page for the guide's query. Used to map neighboring intents a comprehensive page should cover, and as candidates for internal links or sibling articles.
- Get Guide SERP Returns the search results page the guide was built from, with the competing URLs and their positions. Used to record what the guide is actually modeled on and to pull competitor pages for further analysis.
Scoring and Analysis
3- Check Content Optimization Scores a piece of text against a guide and returns semantic optimization and over-optimization ratings along with the terms that are missing, well covered or overused, color coded by health. Used as the pre-publication check on every draft, and the action that makes both directions of the problem visible instead of only the easy one.
- Suggest Entities Extracts the named entities a piece of content should mention to read as authoritative: people, places, brands and concepts. Used alongside Check Content Optimization, since entity coverage is a distinct signal from keyword coverage and a page can pass one and fail the other.
- Explore Questions Returns the questions searchers ask around a piece of content without needing a guide first. Used to mine an existing page for FAQ material or follow-up article ideas when the starting point is arbitrary text rather than a query.
Groups and Cocons
8- Create Group Creates a group to hold related guides. Used before generating the guides for a topic cluster, since the group is the unit a cocon is computed from.
- List Groups Returns the account's groups a page at a time. Used to resolve a group ID before adding guides or building a cocon, and to audit which clusters already exist.
- Delete Group Permanently deletes a group. The guides inside survive, but the grouping and any cocon computed from it are lost with no undo, so a flow should confirm the group ID before calling it.
- Add Guide To Group Attaches an existing guide to a group so it takes part in that group's cocon. Used to reorganize a cluster after the fact, with a cocon rebuild required for the change to take effect.
- Build Cocon Triggers computation of the semantic cocon for a group: the internal-linking structure connecting its guides into a topic silo. Used after adding or removing guides, ahead of reading the result.
- Get Cocon Returns the computed cocon for a group, showing how the guides relate and which pages should link to which. Used to hand a CMS a whole cluster's internal-linking plan in one pass.
- Get Cocon Questions Returns the "People Also Ask" questions aggregated across every guide in a group's cocon. Used to find coverage gaps that span the cluster and would be invisible from any single guide.
- Get Cocon Word Cloud Returns the weighted vocabulary across a group's cocon with relative importance. Used to set an editorial glossary for a cluster and to check a set of drafts collectively covers its language.
Generation
7- Generate Text Continues a passage of text with AI at a chosen optimization level, without tying the output to a guide. Used for free-form continuation, with temperature, word cap and stop words controlling the result.
- Generate Optimized Text Generates text steered by a specific guide so the output already carries the vocabulary that query needs. Used as the highest-leverage writing call, and the one whose output should always be scored by a run of Check Content Optimization that an editor sees.
- Generate Full Article Runs the automatic writing mode against a guide to produce a complete article from a title or short brief, handling structure and vocabulary itself. Used as a first draft an editor then tightens, never as a finished page.
- Generate Outline Produces a heading structure for an article based on a guide. Used as the skeleton before writing, or to audit whether an existing article's structure matches what currently competes for the query.
- Generate Questions Section Writes a question and answer section against a guide, drawing on the questions searchers ask for that query. Used to add an FAQ block that also gives the page a chance at FAQ rich results.
- Rephrase Text Rewrites a passage while preserving its meaning, steered by a guide so the right vocabulary survives the rewrite. Used to refresh dated copy, remove duplication between similar pages, or lift a weak section without rewriting the whole article.
- Translate Text Translates a passage into one of nine supported languages, with no guide involved. Used as the first step when localizing an article, before generating a guide for the target market and scoring the translation against it.
Account
2- Get Generation Credits Returns the remaining allowance for the AI writing features. Used before a batch of generations, since generation is metered separately from guides and a run can otherwise fail partway through.
- Get Account Status Returns the account's guide allowance and plan information. Used together with Get Generation Credits for the full picture before a large run, because the two meters are independent.
Frequently Asked Questions
What can FlowRunner do with YourTextGuru?
FlowRunner agents can run Create Guide, Get Guide, and List Guides in YourTextGuru, plus 23 more actions.
Does connecting YourTextGuru to FlowRunner require OAuth?
No. YourTextGuru connects to FlowRunner with an API key, no OAuth flow required.
Can YourTextGuru trigger a FlowRunner workflow automatically?
YourTextGuru doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger.
Start building with YourTextGuru
$100 in credits. No card required. Connect in minutes.