Local First
Scriplume is being designed around local project storage so writing does not require placing the only working copy inside a hosted application database.
Product privacy principles
Scriplume is being designed so writers can understand where their work lives, when a network is used, and what crosses the AI boundary.
Scriplume is being designed around local project storage so writing does not require placing the only working copy inside a hosted application database.
Local storage does not automatically mean local AI. Model providers are treated as separate, configurable services.
When a remote AI service is configured, the goal is to make network use understandable and send only the context required for the requested task.
The architecture is intended to support local model providers so writers can eventually keep AI-assisted workflows entirely on their own computer.
Manuscript content, biographies, research, and creative assets should never be treated as ordinary diagnostic data.
Writers should be able to retain, back up, export, and recover their own projects without depending on Scriplume’s servers.
Important note: These principles describe the intended product architecture during development and should not be treated as a final legal privacy policy.