Is there a Claude Code skill that runs my own AI workflows?
Yes, Wireflow's workflow runner skill for Claude Code runs your own AI workflows from a plain request, once or per CSV row, after an estimate and your yes.
Yes, Wireflow's ai-workflow-runner skill runs workflows you have already built in Wireflow from a plain request in Claude Code, once or once per row of a CSV. It finds the workflow in your account, reads the inputs it accepts, estimates the credits from the models inside it and waits for your yes, then runs it, polls until it is done, and saves every output to ./wireflow-outputs/ with a runs.csv log.
There is no fixed app behind this skill. It works on your own workflows through the Wireflow MCP connector's list_workflows, get_workflow and run_workflow tools, with any image, video, audio or text workflow you own. Without the connector, it explains how to connect and what it would do.
Skill files and examples are in the skill's folder on GitHub.
How do I install the workflow runner skill in Claude Code?
Install the Wireflow plugin, which adds every Wireflow skill and the Wireflow connector:
/plugin marketplace add wireflowINC/skills
/plugin install wireflow-skills@wireflow
Claude Code shows the plugin's details first, so you can review it and choose its scope. Then run /mcp, choose wireflow and sign in with your Wireflow account. The skill shows up as /wireflow-skills:ai-workflow-runner, and Claude also picks it up from plain requests.
To add only this skill, use the skills CLI, or copy the skills/ai-workflow-runner folder into ~/.claude/skills/. Then connect Wireflow once and run /mcp to sign in. The read and run permissions a new connection gets by default are enough, and there is no API key to copy.
npx skills add wireflowINC/skills --skill ai-workflow-runner
claude mcp add --transport http wireflow https://www.wireflow.ai/api/mcp
How do I run one of my workflows from Claude?
Name or describe the workflow and give the new input:
Run my "GPT Image 2.5 Flare Text to Image" workflow with this prompt:
a speckled sage-green ceramic teapot on pale stone, soft light.
The skill confirms which Wireflow account is connected, lists your workflows and matches the request against their names and descriptions. If two or more fit, it shows the top three and asks which one, and if none fit, it says so instead of guessing. It then reads the workflow's inputs and tells you which ones will keep their saved defaults, because a stale saved prompt is a common surprise.
In the skill's recorded test on 8 October 2026, a fresh Claude Code session found the workflow with list_workflows, read its prompt input with get_workflow, ran it once with run_workflow and saved the image plus runs.csv, for 5 credits.
Can it run a workflow over a spreadsheet?
Yes, it maps each CSV column to a workflow input, checks every row and quotes the whole batch before anything runs. After your yes, it runs once per row with at most 3 runs at a time and records each row's status as it finishes. Failed rows can be run again with a new quote and a new yes.
Use my thumbnail workflow for every title in titles.csv.
Which Wireflow tools does the skill use?
Seven tools from the Wireflow MCP connector, and only run_workflow spends credits.
| Tool | What the skill uses it for | Spends credits |
|---|---|---|
whoami |
Confirm which Wireflow account is connected | No |
list_workflows |
Find candidate workflows by name | No |
get_workflow |
Read a workflow's nodes and the inputs it accepts | No |
list_models |
Look up current prices of the models inside it | No |
get_credit_balance |
Check the balance before and after | No |
run_workflow |
Start one run and get an executionId |
Yes |
get_execution |
Poll the status and collect output URLs | No |
Each tool is described on the MCP connector page.
How much does it cost to run a workflow this way?
The skill itself spends nothing, and each run costs whatever the models inside your workflow cost.
| What | Credits |
|---|---|
| Finding the workflow, reading its inputs, the estimate | 0 |
| Each run | Whatever your workflow's models cost |
| The recorded one-image test run | 5 |
Before it asks, the skill estimates the cost from the current price of each model in the workflow, then shows the estimate, the number of runs, the total and your balance. Afterwards it reports the real cost as your balance before minus after. Workflows with many paid nodes can cost far more than one image.
Runs need credits on your Wireflow account. The Free plan's monthly credits are shown on the pricing page, and paid plans start at $36 a month for 2,660 credits (pricing).
Will it change or build my workflows?
No, it only fills in inputs and runs workflows in your account, or ones shared with you to edit, and it never changes them. If a workflow needs edits, it tells you what to change on the canvas.
To build workflows from Claude with an API key, use the builder skill on the agent skills page. To run workflows from your own code, see Run Workflows.
Does the workflow runner skill work in Cursor, Codex or Claude Desktop?
It is built for Claude Code, and the skills CLI also installs it in Cursor, Codex and other agents that read SKILL.md. Running a workflow then needs that agent connected to the Wireflow MCP server, and the skill's README says Wireflow has not tested those agents itself. In Claude.ai and Claude Desktop, add https://www.wireflow.ai/api/mcp as a custom connector; saving files needs a shell, so there you get the output links instead.
Connection steps for each agent are in Use Wireflow from your agent.