Getting Started
Maestrly is a desktop workspace for running AI-assisted work across chats, agents, tools, and projects. It is local-first, but not local-only: project coordination happens on your machine, while AI inference goes directly from the app to the provider you choose.
This guide takes you from installation to a reviewed first result.
1. Obtain the source preview
Maestrly App is open source under the MIT License and is a 0.x source preview. Official binaries, when available, are only assets on tagged GitHub Releases. A source build or CI package is not a production release. There is no automatic update channel; check Releases manually.
For a source build, follow the prerequisites and commands in the README. Open the app without a Maestrly account: there is no Maestrly login, hosted backend, license service, telemetry, or automatic diagnostic upload. Local workspace features do not require a provider account.
2. Connect an AI provider
Open Settings, choose a supported provider, and follow its connection flow. You need your own provider account, subscription, or API credentials. Maestrly is not affiliated with the providers and does not resell their services.
Provider authentication (including a provider’s Google or GitHub sign-in, if offered) is separate from Maestrly; it does not create a Maestrly account.
Before continuing:
- confirm the provider connection reports as ready;
- choose a model available to your account and region;
- review the provider's pricing and data controls.
Model names and availability can change independently of Maestrly. If a model is missing, verify access in the provider's own console.
3. Add a project
Select Add project and choose a local folder. For code work, choose a Git repository so Maestrly can isolate work, show changes, and support review. Repositories and worktrees remain at their existing filesystem paths.
4. Start your first chat
Create a chat inside the project, then choose:
- the provider and model;
- an agent, if you want a reusable set of instructions;
- the execution mode—your current checkout or an isolated Git worktree;
- any tools or skills the task actually needs.
Start with a bounded request: state the outcome, relevant files or constraints, and how the result should be verified. The prompt and selected context are sent directly to the chosen AI provider for inference under that provider's terms.
5. Review before delivery
Watch tool calls, answer approval requests, and inspect the resulting diff. Run the relevant checks before committing, merging, or deploying. AI output can be incomplete or incorrect; you remain responsible for actions and delivered changes.
6. Build the workflow
- Chat and models explains model selection and request failures.
- Agents and skills turns useful instructions into reusable setups.
- Projects and Git covers local checkouts, worktrees, and delivery.
- Usage and costs explains provider billing and limits.
Need help? Visit Support.