VibeMarketer

Technical founders often know their product but lack a repeatable distribution workflow. Marketing context gets lost between prompts, drafts, approvals, and publishing tools.

VibeMarketer turns a product URL into a controlled campaign loop:

Context → Plan → Create → Approve → Publish → Learn

It extracts attributable product evidence, builds persistent brand memory, plans a seven-day campaign, generates channel-specific drafts, and routes every publish action through human approval. Provider connectors must confirm the external action before the product records it as published; unavailable providers remain visibly unavailable instead of returning fake success.

The differentiator is continuity and accountability. Campaign outputs stay tied to the source evidence and brand decisions that produced them, while execution records make the final action inspectable.

The application uses Next.js, a typed agent workflow, Supabase authentication and Postgres with row-level security, plus provider-specific publishing connectors. It supports a judge-verifiable path from URL ingestion to an approved, provider-confirmed action.

Commercial traction, revenue, and Google Cloud usage require separate evidence; they are not inferred from generated content or a rendered interface.

Built With

• ai
• composio
• e2b.dev
• firecrawl
• gemini
• gemini-api
• next.js
• postgresql
• supabase
• tavily
• typescript
• vercel
• vertex