Documentation
Everything about Yuli: installation, the IDE, the agent, MCP and ACP, quality, governance and security.
Getting started
- What is YuliA complete framework for developing with AI without losing control: IDE, agent, metrics, methodology and quality.
- Install YuliDownload the IDE for free for 64-bit Windows, macOS, or Linux and open it for the first time.
- Your first projectOpen a folder, let Yuli index it, decide on trust and run the application for the first time.
- Connect an AIChoose between the subscription you already pay for or your own API key. The IDE and its rules stay the same.
The IDE
- The interfaceTitle bar, rails, chat, editor and bottom panel. Everyday tools in view; everything else, one click away.
- Editor and languagesHighlighted languages, language servers, the native Java engine and the everyday editing tools.
- Code intelligencePrecise navigation, references, hierarchies, refactorings with preview, and symbol search.
- Run and debugBuilds with Maven, Gradle, Ant, npm or Cargo, application servers, JDWP, CDP and PL/SQL debuggers, hot swap and JVM diagnostics.
- DatabaseDatabase client for Oracle, SQL Server, MySQL, MariaDB, MongoDB and SQLite, with a SQL and PL/SQL console, execution plans and code ↔ database navigation.
- Web StudioThe front-end workspace. Design in Design Studio, see the app without running it, mock APIs and bring up the whole system with Compound.
- REST Client and diagramsTest your APIs next to the code that calls them, and understand the system with architecture, flow, business, and database diagrams.
- Git and CI/CDA full Git client, conflict resolution, and Jenkins, GitHub Actions, GitLab CI and CloudBees pipelines inside the IDE.
- Containers and remote developmentDocker, Compose and Kubernetes from the IDE, and working on WSL, SSH or containers as if they were local.
The agent
- Ask for a taskHow to work with the agent from start to finish without losing control.
- Missions and phasesEvery large job is a mission with an 8-phase lifecycle, an execution plan, and a goal that never gets lost.
- Review changesThe AI changes window: nothing reaches the disk without your review, and everything can be undone with checkpoints.
- Agent teamFor large jobs, a lead agent distributes tasks among parallel sub-agents, integrates the result, and has it reviewed.
- Unattended modeThe agent asks only during analysis, decides and records the rest, and carries on by itself until it finishes, without bypassing any limit.
- Memory, skills and rulesHow to give the agent lasting context with project memory, skills, prompts, hooks and rules, and how it learns from its mistakes.
- Models and providersSupported providers in Platform mode, local models, extended reasoning, context, caching, usage and budgets.
MCP and ACP
- Yuli as an MCP serverHow yuli-mcp exposes the whole IDE to your terminal agent, how it connects and what happens when Yuli is closed.
- Claude CodeRun your own Claude Code inside Yuli, with your account, connected to the whole IDE and under the same rules.
- Kiro, Codex and AntigravityYuli as both an ACP client and an MCP server, so you can work with your own Kiro, Codex or Antigravity inside the IDE.
- Yuli toolsThe catalog of yuli_* tools your agent receives, by family and with usage examples.
Quality and governance
- Quality contractQuality rules live in a versioned file in the repository, not in the prompt, and apply equally to people and agents.
- Measuring against N−1How Yuli measures quality, always compares it with the previous version and makes the agent improve before testing.
- Gate and Governance CenterA single gate enforces the contract on every commit, push and pipeline, and the Governance Center brings the whole lifecycle together.
- The .ai folderWhat Yuli stores inside your project, what is versioned with the team and what stays only on your machine.
Security
- SecurityLocal first and deterministic rules. What Yuli protects, what leaves your machine and what you decide.
- Sandbox and permissionsWhere the agent can work, what it does without asking and what always waits for your approval.
- Secrets and privacySensitive data never leaves your machine. How Yuli blocks, redacts and restores secrets, and how it defends against prompt injection.
- Secure databaseYou decide, connection by connection, what the agent can do in your database. Nothing writes without your approval.
- Security CenterLocal analysis of dependencies, code, memory leaks and race conditions, without sending your code anywhere.
Reference
- Keyboard shortcutsYuli's shortcuts by category, and how to reassign them.
- LanguagesYuli speaks 8 languages, follows your system's language and can be corrected or extended without rebuilding.
- ArchitectureHow Yuli is built: a native Rust engine, an Electron interface with TypeScript and React, and the language model as the layer that reasons.
- Frequently asked questionsShort answers to the most common questions about Yuli, AI, security and cost.
- TroubleshootingWhat to do when something doesn't work: startup, indexing, AI, Claude Code, Git, database and build.