> ## Documentation Index
> Fetch the complete documentation index at: https://velurahq.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.
# Workflows: automate content research and publishing
> Build visual pipelines that chain research, AI drafting, review gates, and CMS publishing into one governed, repeatable content system.
VeluraHQ Workflows let you turn your best content process into a repeatable, governed system. Instead of coordinating research in one tab, drafting in another, and publishing manually from a third, you connect every step on a visual canvas — with live data, Brand Kit context, human review gates, and direct CMS publishing built in. Once a workflow exists, anyone on your team can run it, inspect every step's output, and trust that brand standards travel with the work.
## How Workflows work
Connect research, drafting, optimization, review, and publishing blocks on the visual canvas. Drag blocks from the sidebar and connect them with lines to define the order of execution.
Choose the AI model for each step (GPT-4o, Claude, Gemini, and more), map input variables from earlier blocks, and attach Brand Kit context so every generation stays on-voice.
Insert human review steps where judgment matters. Assign an owner from your workspace members; the workflow pauses at that gate until they approve or request changes.
Run the workflow on sample inputs and inspect each block's output. Adjust prompts and conditional logic until quality is consistent across different inputs.
Run manually, schedule with Pulse, trigger via API, or execute across all rows in a Sheet. The same workflow handles one-off runs and large-scale bulk operations.
## Trigger methods
Trigger immediately from the workflow builder or your workspace dashboard. Useful for one-off jobs and testing runs before you automate.
Set daily, weekly, or monthly cadences so recurring work — refreshes, monitoring, archive maintenance — happens without manual prompts.
Point a Sheet column at any workflow to execute it across every row in bulk. Results land directly in the adjacent columns of the same Sheet.
Launch a workflow from your CRM, data warehouse, or any external system using the VeluraHQ API. Pass dynamic inputs at call time.
Workflows triggered via API consume credits identically to manual and scheduled runs. Implement rate limits in your external system to avoid unexpected credit depletion.
## Workflow blocks
Every block in the workflow builder serves a distinct purpose in the content pipeline. Mix and match them to match your process.
**Research blocks**
Pull live, current data into your workflow before any generation step runs.
* **Web search** — Query the live web for a topic, keyword, or competitor URL.
* **URL scrape** — Extract structured content from any page.
* **Semrush keyword data** — Pull keyword volume, difficulty, and intent data.
* **Ahrefs backlink data** — Surface backlink profiles and referring domain data.
* **Google Search Console sync** — Bring in live impressions, clicks, and position data for pages you own.
**AI/LLM blocks**
Generate text at any step. Choose the model that fits the task — GPT-4o, Claude, Gemini, or Perplexity — and configure the prompt and context per block. Different steps in the same workflow can use different models.
**Logic blocks**
Branch the workflow based on conditions. Use if/else branching by content type, quality score, or the output of any earlier block to send high-confidence outputs on a fast path and edge cases to a deeper review path.
**Review blocks**
Insert a human review gate between any two blocks. Assign an owner, set an optional SLA, and the workflow holds at that point until a person acts.
**Publishing blocks**
Push approved drafts directly to your CMS. Connect to WordPress or Webflow and map workflow output fields to CMS fields — no copy-pasting required.
**Data blocks**
Read from and write to Sheets, and pass variables between steps using liquid-style placeholders. Use these blocks to hydrate a workflow from a spreadsheet or to write results back after a run.
## Brand Kit and quality scoring
Every workflow can pull your Brand Kit context at each generation step. Voice rules, terminology lists, style guidelines, and tone settings travel with the job — not just into the first block, but into every block that produces output.
Before a draft reaches human review, VeluraHQ can automatically score it across three dimensions:
* **Brand alignment** — How closely the output matches your voice, tone, and terminology rules
* **Structure quality** — Heading hierarchy, paragraph length, and readability
* **Citation readiness** — Presence of verifiable claims that AI answer engines can cite
Set a minimum quality score threshold on your human review gate. Drafts that clear the threshold continue directly to publish; only outputs that fall below it pause for a reviewer. On high-volume Sheet runs, this significantly reduces review load without sacrificing control over edge cases.
## Human review gates
Review gates are the control points that keep people in charge of the decisions that matter. When a workflow reaches a gate, it pauses and notifies the assigned owner. Nothing moves to the next step — including CMS publishing — until that person acts.
The reviewer can approve the draft, request changes, or reject the run entirely. If they request changes, their feedback is attached to the same run record and passed as context into the next generation attempt, so the model has the reviewer's specific notes — not just a generic retry.
Because feedback travels with the run, every revision attempt is traceable. You can see what was requested, what was generated in response, and when approval was granted, all in one place.
## Workflow templates
VeluraHQ includes pre-built templates for the most common content operations so you can start from a proven structure rather than a blank canvas:
* **Content Refresh** — Detect decay, pull live SERP data, and regenerate updated sections.
* **Article Generation from Brief** — Take a structured brief through research, drafting, AEO scoring, and publish.
* **Programmatic Page Creation** — Generate templated landing pages at scale with Brand Kit rules intact.
* **Meta Description Optimization** — Rewrite meta descriptions across a URL list in one bulk run.
* **Internal Linking Pass** — Add internal citations across your content library from a single workflow.
Start from a template and customize any block — swap the model, adjust the prompt, add a branching condition, or change the review owner — to match your exact process.
## FAQ
Every run can use Brand Kit context, structured rules, automated scoring, and human approval gates before production outputs are released. Brand rules live in the system itself, not in a human's memory, so they apply consistently whether you run the workflow once or a thousand times.
Yes. Use the same visual builder for article generation, bulk refreshes, meta updates, and any other repeatable content task. The trigger method and block configuration differ, but the workflow structure is the same.
Yes. Research blocks pull live data from Semrush, Ahrefs, and Google Search Console. Publishing blocks push approved assets to WordPress and Webflow. The API trigger accepts calls from any external system — CRMs, data warehouses, internal tooling, or event-based automation platforms.