# Duvo > Build, run, and monitor autonomous AI agents that execute real work. - [User Guide (162 pages)](https://docs.duvo.ai/_llms/user-guide.md): Documentation for User Guide. - [API Reference (408 pages)](https://docs.duvo.ai/_llms/api-reference.md): Documentation for API Reference. ## MCP ### Overview - [How MCP works on Duvo](https://docs.duvo.ai/mcp/overview.md): Understand the two directions Duvo speaks MCP — driving Duvo from an MCP host like Claude Desktop, Claude Code, Cursor, or ChatGPT, and bringing your own MCP servers into Duvo as Connections. ### The Duvo MCP server - [Connect to the Duvo MCP server](https://docs.duvo.ai/mcp/duvo-mcp-server.md): Connect Claude Desktop, Cursor, or ChatGPT to the Duvo MCP server. Every Public API endpoint becomes an LLM-callable tool. - [Available MCP Tools](https://docs.duvo.ai/mcp/available-tools.md): Complete reference for all tools exposed by the Duvo MCP server. - [Embedding Duvo as a Callable Tool in Your AI System](https://docs.duvo.ai/mcp/embedding-duvo.md): Register Duvo Agents as callable tools your own AI agent or copilot invokes through the Duvo MCP server — your agent orchestrates, Duvo handles the operational task. ### Bring your own MCP - [Connect a custom MCP to Duvo](https://docs.duvo.ai/mcp/custom-mcp-servers.md): Bring your own MCP server into Duvo as a Connection. Let agents call tools from internal systems, legacy platforms, or custom APIs alongside built-in connections. - [Building an MCP Server for Duvo](https://docs.duvo.ai/mcp/building-mcp-servers.md): Build a custom MCP server that Duvo can connect to. Covers transport, authentication, tool design, and deployment requirements. ## CLI ### Get started - [Duvo CLI](https://docs.duvo.ai/cli/index.md): Install and use the Duvo CLI to manage Agents, start Runs, move Files, and drive the Duvo API from your terminal or CI/CD pipelines. - [Profiles and Teams](https://docs.duvo.ai/cli/profiles-and-teams.md): Manage stored CLI profiles, switch between teams, and inspect team membership with the Duvo CLI. ### Clarity Processes - [Clarity CLI](https://docs.duvo.ai/cli/clarity.md): Use the Duvo CLI to inspect, update, and manage Clarity processes from your terminal. ### Agents - [Managing Agents](https://docs.duvo.ai/cli/managing-assignments.md): List, create, update, and delete Agents, organize them into folders, and manage their Revisions and memory with the Duvo CLI. - [Starting Runs (CLI)](https://docs.duvo.ai/cli/running-jobs.md): Start Runs, check status, stream messages, respond to human-in-the-loop requests, and stop active Runs with the Duvo CLI. - [Scheduling and Triggers](https://docs.duvo.ai/cli/scheduling-and-triggers.md): Start Runs automatically on a schedule or in response to events and Cases, using the Duvo CLI. ### Connections, Skills & Secrets - [Managing Connections](https://docs.duvo.ai/cli/managing-connections.md): List and inspect your authorized Connections, browse your team's enabled Connection types, and start OAuth flows with the Duvo CLI. - [Managing Files, Skills, and Plugins](https://docs.duvo.ai/cli/managing-files-and-skills.md): Move documents in and out of your team's file library, create and manage Skills, and browse Plugins from the Duvo CLI. - [Secrets and Logins](https://docs.duvo.ai/cli/secrets-and-logins.md): Store env-var secrets and logins, and attach them to an Agent Revision so Runs can use them, with the Duvo CLI. ### Queues - [Cases and Queues](https://docs.duvo.ai/cli/cases-and-queues.md): Inspect Case history, bulk-reprocess Cases on an Agent, and manage labels with the Duvo CLI. ### Automation & advanced - [Scripting and CI/CD Patterns](https://docs.duvo.ai/cli/scripting-and-ci.md): Run the Duvo CLI in scripts and CI/CD pipelines with non-interactive authentication, JSON output, and worked automation examples. - [Advanced CLI Commands](https://docs.duvo.ai/cli/advanced.md): Lower-level Duvo CLI commands for sandboxes, revision integrations, raw API access, and deprecation notices. ## Best Practices ### Overview - [Best Practices](https://docs.duvo.ai/best-practices/index.md): Recommendations and proven patterns for getting the most out of Duvo Agents. - [Tips and Patterns](https://docs.duvo.ai/best-practices/tips-and-patterns.md): A cross-referenced index of the practical guidance across the Duvo docs, grouped by what you're trying to do. ### Building Agents - [Writing Effective AOPs](https://docs.duvo.ai/best-practices/writing-effective-aops.md): The AOP is where most of an Agent's quality lives. Write clear, testable instructions that produce reliable, predictable Runs. - [Choosing How Your Agent Starts](https://docs.duvo.ai/best-practices/choosing-how-runs-start.md): Manual Start Work, schedules, triggers, and Queues each fit a different shape of work. The start mechanism decides latency, retries, and what happens when Runs overlap. - [Designing Work Around Queues and Cases](https://docs.duvo.ai/best-practices/queues-and-cases.md): Structure producer and consumer Agents so every Case ends somewhere definite, waiting periods work with the platform, and nothing silently disappears. - [Connecting Agents with Queues](https://docs.duvo.ai/best-practices/connecting-agents-with-queues.md): One Agent that does everything gets worse as the workload grows. Split the process into stages connected by Queues, and every item gets its own Run — the thousandth processed as reliably as the first. - [Where Knowledge Belongs: AOP, Skills, Files, or Memory](https://docs.duvo.ai/best-practices/where-knowledge-lives.md): Four places to teach an Agent something, each with different scope. - [Giving Recurring Runs a Memory](https://docs.duvo.ai/best-practices/recurring-runs-memory.md): Recurring Agents that pick from a pool will repeat themselves unless you design where their memory lives. Externalize state between Runs, exclude at the source, and validate before writing. ### Skills & Scripts - [Let a Skill Do the Math](https://docs.duvo.ai/best-practices/let-a-skill-do-the-math.md): When results must be exact and auditable, package the calculation as a script inside a custom Skill. The Agent orchestrates; the engine computes. - [Wrap a REST API in a Skill](https://docs.duvo.ai/best-practices/wrap-a-rest-api-in-a-skill.md): When there is no native Connection for a system but it has an HTTP API, package the call as a script inside a custom Skill. The Agent decides what to ask; the script makes the request the same way every time. - [Browsing That Survives UI Changes](https://docs.duvo.ai/best-practices/browsing-that-survives-ui-changes.md): Browser Agents break when they depend on where things are instead of what things are. How the Agent actually sees a page, what makes a recorded browsing Skill replayable, and the habits that keep browser work running after the next redesign. ### Connections & Access - [Connections and Logins That Don't Break](https://docs.duvo.ai/best-practices/reliable-connections.md): Most "my agent stopped working" reports trace back to access. Set up Connections, Logins, and Secrets so they survive teammates, departures, and expired credentials. - [One Builder, Many Users](https://docs.duvo.ai/best-practices/sharing-access-across-a-team.md): An Agent that works perfectly for its builder can silently fail for everyone else. The two access gates behind "it works for me", how to share Logins and Secrets properly, and how to prepare an Agent for whole-team use. ### Operating in Production - [When the Agent Should Ask, Not Act](https://docs.duvo.ai/best-practices/when-to-ask-a-human.md): Approval gates fire only where the AOP asks for them, the request goes to one specific person, and there is no timeout — a well-designed gate names the action, reaches the right inbox, and is worth waiting for. - [Make Every Run Auditable](https://docs.duvo.ai/best-practices/auditable-runs.md): When someone asks "what did the Agent do, and was it right?", the answer should be a link, not an investigation. What Duvo records for every Run, how automatic Evaluations work, and what you must design into the AOP yourself. - [Iterating on a Live Agent](https://docs.duvo.ai/best-practices/iterating-on-a-live-agent.md): Editing the active revision changes production immediately. Fork a draft, test it on real work from the builder, make it active deliberately, and know exactly what switches over the moment you do. ## Changelog - [Changelog](https://docs.duvo.ai/changelog.md): Weekly highlights of Duvo's user-facing updates. ## OpenAPI Specs - [mintlify](https://api.duvo.ai/v2/documentation/mintlify.json) > The links below point to documentation indexes. Follow each `/_llms/` index recursively until you reach documentation pages. ## Indexes - [User Guide (162 pages)](https://docs.duvo.ai/_llms/user-guide.md): Documentation for User Guide. - [API Reference (408 pages)](https://docs.duvo.ai/_llms/api-reference.md): Documentation for API Reference. - [API Reference / Clarity (64 pages)](https://docs.duvo.ai/_llms/api-reference/clarity.md): Documentation for API Reference / Clarity. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.