Paste a client transcript/offer and details, then Copy the AI prompt into your AI chat and paste the JSON back — or use the optional Live Generate below.

For the best board, paste this prompt into one of the latest top models — e.g. Claude Opus 4.8, Gemini 3.1 Pro, GPT-5.x or Mistral Large — then paste the JSON it returns into the Spec editor above.

Runs the same prompt against your own AI provider. Everything is optional — the app works fully offline without it. Choose a provider and paste its API key; keys are stored only in this browser and sent directly to the provider you pick, over HTTPS. Nothing else leaves your machine. Each provider remembers its own model + key, so you can switch between Gemini and Claude without re-entering anything.

Save the facilitators and workshop-title templates you reuse. Facilitators show up in the multi-select above (pick one or more); titles appear as suggestions in the Workshop title box. Each title is just the last part (e.g. “GTM Kick-off”) — the “{Client Name} × CxLEVER —” prefix is added automatically.

House templates: type a name and click Save to store the current board as your own reusable template, then pick it here and Load it later. (The session length is set by Workshop Duration in the Generate panel above — no need to choose it here.)

Versions: Save snapshot stores a dated copy of the current spec (with its readiness score). For any snapshot, Diff shows exactly what changed vs the board you have open now (added / removed / re-typed sticky notes, zone and title changes) — handy to compare the board before vs after the client call — Restore loads it back, and the trash icon deletes it.

Scores the board against the CxLEVER method — hypotheses to validate, blue prompts for client input, risks visible, owners & dates locked — before the session.

Self-test of the app's dependencies and current config. No data is sent unless you click "Test connection" above.