Point Claude, Cursor, or your own setup straight at your Slima project, and work with your whole book from wherever you write.
MCP is a shared language for tools to talk to apps. Slima speaks it, so your editor reaches the same full-book context Slima keeps, not a pasted snippet.
Claude, Cursor, Gemini CLI, and anything that speaks the open standard.
Your tools get the same full-book context Slima keeps, not a fragment.
Draft, edit, and organize from your editor; it lands in your project.
Point a hosted client at your Slima project and authorize it once. Your book is there, ready to read and write, no install needed.
Run the connection on your own machine and your editor talks to your project directly. The same tools, right inside where you build.
Slima exposes your project as a clean set of tools. Your client lists them, then reads chapters, drafts new scenes, and keeps everything in your project.
List and read chapters, with the full-book context Slima already holds.
Draft and edit chapters straight from your editor; it lands in your project.
Pull characters, notes, and timelines so your tool never loses a thread.
See snapshots and restore points, so nothing you write is ever lost.
Each connection runs on a token you create and scope to one project. See every authorized app in one place, and revoke any of them whenever you like.
Connecting a tool never opens your work to anyone else. Your book stays yours, encrypted, and never used to train models.
Nothing is shared until you authorize a client and scope its token.
Pull any token and that client loses access at once.
Your project is encrypted and never used to train anyone's models.
Start free and connect your setup in minutes. Your work is never used to train models.