Build and QA customized campaigns from proven templates.
Start in the Workspace, choose a funnel flow, wire in your products, preview every page, and ship with validation already baked in.
Every template family is a runtime contract — pages, slots, and cart wiring the Campaigns SDK understands. Pick the one shaped like your offer.
The fastest way through the loop isn’t by hand. Campaigns OS ships agent skills and machine-readable contracts, so an AI agent can scaffold, build, and QA a campaign end to end — you review and launch.
Five lifecycle skills ship with the public toolkit for Claude and Codex — orchestration, setup, build, polish, and QA — plus agent context for Copilot and Cursor.
Every artifact is machine-readable — CampaignSpec, Build Packet, QA verdict. Agents consume the same contracts you review.
llms.txt points your agent at the templates, patterns, and toolkit — it can learn the system without a walkthrough.
Every CampaignSpec is validated against the schema — products, prices, and page wiring.
Every page renders in a real browser before review. Broken markup never reaches QA.
QA places real test orders through the checkout — card entry, upsells, receipt — using global test cards.
When a campaign passes, you get a handoff packet — not a promise.
Open the Workspace, load your campaign, and leave with a build that has already been verified.