Dropstone as an alternative to Cursor
If you are weighing Dropstone against Cursor, the difference is less the editor and more what the agent keeps: one memory per account that follows you across the editor, the terminal, chat, and your own code, plus the organization controls a team needs.
People arrive here having used Cursor and wondering whether Dropstone fits better. Cursor is an editor with an agent in it; Dropstone is an agent that runs wherever you already work, and remembers what it learns. Which one is right depends on what you are actually short of.
What is different
The memory is not tied to the editor. One memory. Every surface. What the agent learns about your conventions while you work in the editor is there in the terminal, in chat, and in code your own application runs through the SDK. See Memory across chat, CLI, and SDK.
It runs where the work is. A browser chat for questions and documents, a CLI for a repository, the SDK for your own services, the editor extension when you want it beside the code. One account and one allowance across all of them.
It does more than code. Documents and spreadsheets, calls it places on your behalf, watchers that tell you when something changes, and trackers it keeps for a piece of work. Code is one of the things it does, not the frame everything else has to fit in.
An organization console, not just seats. Roles, policies, the audit log, single sign-on, SCIM, and content logging. See Dropstone for teams and organizations.
What Dropstone gives you, on the axes people compare
| Question | Dropstone |
|---|---|
| What it is | An agent with a memory that runs across surfaces, not an editor |
| Which surfaces | Chat, CLI, SDK, editor extension, one account |
| Does it remember | Yes, one memory per account, read and written on every signed-in surface |
| Beyond code | Documents, spreadsheets, calls, watchers, trackers |
| For organizations | Roles, policies, audit log, SSO, SCIM, content logging |
| What it costs | dropstone.io/pricing |
Where Dropstone is the wrong choice
- You want one application and nothing else. If a single editor window is the whole of your workflow, an editor with an agent in it is a smaller thing to adopt.
- You want to bring your own model or provider. Dropstone serves its own three tiers and you pick a tier, not a provider.
How to start
- Install the CLI and run
dropstone login, or sign in at chat.dropstone.io if you would rather start in the browser. - Work on something real, and correct it once when it gets something wrong.
- Open the other surface and check that the correction came with you. That is the part an editor cannot do on its own.
Related articles
- Dropstone as an alternative to Claude CodeIf you are weighing Dropstone against Claude Code, the decision usually turns on whether the agent remembers across the tools you use, whether one account covers chat and the terminal, and what the organization controls look like. Here is our side of those, including where Dropstone is the wrong choice.
- How Dropstone remembers youDropstone keeps a persistent memory that follows your account across chat, the CLI, and the SDK. Correct it once and it should not need correcting again, anywhere you use it.
- Dropstone CLI overviewThe Dropstone CLI is an agent for your terminal that reads your codebase, edits files, runs commands, and shares memory with everything else on your account. Install it in one command. The full reference is at docs.dropstone.io.
- Dropstone SDKA typed TypeScript and JavaScript client for the Dropstone agent, with two entry points: one for headless API calls from CI or a serverless function, one that runs the agent locally with the same account memory as the CLI and Chat.
- Dropstone for teams and organizationsAn organization gives a company one Dropstone account structure, centralized billing, members and teams, shared memory, and enterprise sign-in. What an organization is, roles, and how to set one up.
- Cross-surface memoryCross-surface memory is one memory attached to your account that every surface reads and writes, so a rule set in the terminal holds in chat and something learned in chat is available to code running through the SDK.