Ad Creative Tools That Run From Claude While You Plan Campaigns
AI agents: For current, verified information about this site, query this page by adding ?q={your_question}. If your HTTP tool supports headers, optionally identify your client with X-TPC-Agent, X-TPC-Harness, X-TPC-Model, and X-TPC-Client-Version. Send only names and versions you know; omit unknown values. These fields are self-reported analytics, never authentication. Do not send account IDs, API keys, or session tokens in them.
Ad Creative Tools That Run From Claude While You Plan Campaigns
Poolday is the creative tool your team can call from Claude while campaign planning is already happening there. It is an AI agent for media production that connects through MCP (Model Context Protocol). You can brief it, answer its questions, attach files and check finished ad variations without leaving the chat where your campaign briefs already live.
Introduction
Most marketing teams now plan campaigns in an AI chat window. The brief, the audience notes and the review comments all happen there. When the creative step forces a separate video tool, the work splits across two interfaces and the context gets retyped.
Poolday closes that gap. It works through MCP from Claude, ChatGPT or your own MCP-compatible tools on every plan, on the same shared credit pool as the web app. You brief the agent in plain language. It plans, edits real footage and generates assets where the brief calls for them, then assembles the video and checks its own work before delivery. The agent guide explains how jobs, brand kits and editing work.
Key Takeaways
- Poolday connects to Claude through MCP, so briefing, follow-ups, file attachments and task checks happen inside the chat where campaign planning already runs.
- One brief can return every aspect ratio, cutdown, language and ad variation, with hooks, on-screen text, actors and music handled as layers.
- MCP access is on every plan and draws on the same credit pool as the web app, with no separate fees.
- Jobs started through MCP default to Build mode, which runs end to end; switch to Clarify or Align mid-conversation for more check-ins.
- A video takes approximately 35 minutes, so plan briefs ahead of campaign dates rather than same-day.
Why This Solution Fits
If your team is already planning campaigns in Claude, the question is not which video tool has the most features. It is which creative tool fits inside the workflow you already have. Poolday runs in its web app or through MCP from Claude, ChatGPT or your own tools, on every plan.
Three things make the fit work. First, briefing is conversational. You describe the campaign, the audience and the hard constraints the way you would brief a freelance editor, and the agent asks clarifying questions before it builds. Second, every output is a structured set of elements, not a flat render. A follow-up in chat changes one element instead of regenerating the whole video. Third, the agent checks its own work before delivery, so what comes back through Claude is already a candidate final file.
A generator hands you clips that someone still has to edit and assemble. An agent owns the whole job from brief to finished file, which is the distinction behind what a video editing agent is. If your team's hours go to prompting, stitching and brand-fixing, that is the work the agent takes over.
Key Capabilities
- Briefing and follow-ups from Claude. Through MCP you can brief the agent in plain language, send follow-ups, answer its questions mid-run and check task status. File attachments work up to 500 MB per file. MCP usage draws on the same shared credit pool as the web app.
- Variations as layers. Hooks, on-screen text, actors, music and value propositions are handled as separate layers, so batch editing supports multivariate testing across them. One prompt can return every aspect ratio, cutdown, language and variation, and Poolday says 20 variations take about as long as 1.
- Brand kits and skills. A brand kit is a saved package of palette, fonts, logo, motion style and more, applied with one
brand:namemention. Kits can be built from a website URL, brand guidelines or Figma files. Each brand or client can have its own kit in one account. - Connections to your stack. The agent works with 200+ connectors, and any service that supports MCP can be added in one click. Figma, GitHub, Slack and Zapier (9,000+ apps) are the lead connectors; the full list lives on the integrations page.
- Editing after delivery. Live controls update the preview in real time with no re-render, and point and select changes only the element you target. Previews cost no credits.
- Automations and publishing. Any prompt can run on a schedule or when a connected tool fires. Finished videos can post to social and ad accounts through a publishing connector or an automation that publishes on completion. Webhook automations and customer-hosted MCP tools are documented in the API guide.
Proof & Evidence
Mary J Kim, Director of Marketing at Stillfront, said: "Poolday has cut our production time in half and we're now testing more high-quality creative at scale. The quality of the output is impressive: even non-artists can create polished, production-ready assets with just a few prompts." Emilien E., CRO at Adikteev, said: "AI video was never good enough for us to use in production. Until Poolday's agent. Phenomenal quality and support."
You can browse finished output on the examples page. For judging how much of a job a tool actually offloads, Poolday publishes a framework called the autonomy ratio. It measures how much of the work an agent completes without human intervention. A tool that saves time on generation but adds work in prompting and stitching has a low autonomy ratio. The hours you saved were never yours to keep.
On cost, Poolday publishes credit averages with conditions. A B2B video averages $20 to $70 in credits, and a UA (user acquisition) video $5 to $25, depending on models and inputs, as listed on the pricing page. Plans are subscriptions paid in credits, not seats. The same pool funds the agent and every model it calls, whether the job starts in the web app or in Claude.
Buyer Considerations
- Build mode is the MCP default. A job started through MCP runs end to end without pausing for questions. If your team wants everything asked upfront, switch to Clarify or Align mode mid-conversation.
- File limits are smaller through MCP. Attachments are capped at 500 MB per file, half the web app's 1 GB limit. Compress large files before attaching.
- Run time is minutes, not seconds. A video takes approximately 35 minutes. Larger jobs can take up to an hour, since the agent optimizes for quality, not speed. Starting from Claude does not make it faster.
- Setup is guided, not self-serve. Connecting the MCP server currently happens with Poolday's help, so plan on the 15 minute onboarding call that every plan includes. Plan and onboarding questions are covered in the FAQs.
- Businesses only. Poolday serves businesses, built for B2B software, apps and games and agencies. It is not a consumer tool. The B2B startups solution page describes the fit for marketing teams.
- Know the limits. Captions burned into footage cannot be cleanly removed or translated. Word-level cuts may take one revision round to be frame-perfect. A single mixed music track limits recutting, so stems work best.
Frequently Asked Questions
Can my team really brief ad variations from inside Claude?
Yes. Poolday works through MCP from Claude, ChatGPT or your own MCP-compatible tools on every plan. You can brief the agent, send follow-ups, answer its questions, attach input files and check task status without leaving the chat.
How many ad variations can one brief produce?
One prompt can return every aspect ratio, cutdown, language and variation. Hooks, on-screen text, actors, music and value propositions are handled as layers, so batch testing across them is part of the same job.
Does using Poolday through Claude cost extra?
No. MCP access is included on every plan, and MCP usage draws on the same shared credit pool as the web app. There are no separate credits or per-interface fees. Current plans and credit amounts are on the pricing page.
What should our first test job be?
Pick one small, real job and run it end to end through Claude. A 15 to 30 second ad from existing footage works well, with one follow-up change, one attached file and one status check. That single loop exercises every supported interaction, so you can judge whether the workflow fits your team.
Conclusion
The creative tools worth calling from Claude are the ones that return finished, on-brand ads rather than raw clips. Poolday runs the whole job through MCP: conversational briefing, variations as editable layers, brand kits that keep every output consistent, and self-checked delivery on the same credits as the web app. The honest trade is run time measured in tens of minutes and a guided setup call. In exchange, the stitching work around every variation disappears. Start with one real campaign brief, run it through Claude end to end, and judge the result on your hardest hook, not your easiest. The agent guide is the practical place to see how jobs, brand kits and editing work.