Six agents inspect your idea from every angle. Market, risk, feasibility, opportunity — a verdict in seconds.
Each agent has a job. They run in parallel where they can, in sequence where they must.
Reads the idea
Studies the market
Finds the risks
Finds the edge
Maps the build
Writes the verdict
Reads the idea
Studies the market
Finds the risks
Finds the edge
Maps the build
Writes the verdict
Minutes, not weeks
Skip the spreadsheet phase. Get a defensible read in under a minute.
Six perspectives
Each agent stress-tests a different axis — no single-angle blind spots.
Built on real data
Live web search, competitor signal, and market context — not hallucinations.
Yours to keep
Your ideas are encrypted at rest. We never train on your inputs.
One general model gives you one perspective. Lens orchestrates six specialized agents that run in parallel — each with its own prompt, tools, and constraints — then synthesizes a verdict. You get conflicting takes resolved, not averaged.
Yes. The Researcher agent uses live web search to surface real competitors, market signal, and trend data. It cites what it finds so you can verify.
Your inputs are encrypted at rest and never used to train models. Analyses are private to your account. You can delete them at any time.
The 0–10 score is a directional signal, not an oracle. Treat it as a starting point — the recommendations and risks matter more than the number.
Lens works best for product, feature, and venture ideas. For pure research questions or essays, you're better off with a chat tool.