> For the complete documentation index, see [llms.txt](https://docs.lleverage.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.lleverage.ai/build-and-improve/overview.md). # Set up a Project Set up a Lleverage Project with the systems, context, automation, operator surfaces and controls required for a defined area of operational work. A Project is the working environment for a defined area of operational work. It can contain an Agent, Skills, Connections, Files, Tables, Knowledge, Workflows, Requests, Monitoring views, operator pages and access controls. Use guided setup when a suitable process template exists. Use manual setup when the Project does not fit an available template or when you need to configure the components directly. ## Guided setup When creating a Project, select a domain and process template. Lleverage creates a setup plan and the Agent works through the required tasks with you. See [Guided Project setup](/build-and-improve/overview/guided-project-setup.md). ## Manual setup For manual setup: 1. Define the Project purpose and operational boundary. 2. Configure Project instructions and Agent behaviour. 3. Connect the systems required for the work. 4. Add maintained context such as Files, Tables and Knowledge. 5. Create or update Skills for recurring Agent work. 6. Build Workflows for stable background execution where appropriate. 7. Configure operator entry points, Requests and Monitoring. 8. Test representative cases before using the Project in production. See [Set up an Agent manually](/build-and-improve/overview/first-agent.md) for the Agent-specific path. ## Project components * **Project description and Agent settings** define persistent Project-level direction. * **Skills** define reusable Agent methods. * **Connections** provide authorised access to external systems. * **Files, Tables, Knowledge and Memory** provide different forms of Project context. * **Workflows** run defined execution paths. * **Requests** collect human decisions or input from running processes. * **Monitoring** shows execution and operational results. * **Members and Channels** determine who can use the Project and where. Keep rules and data in the narrowest durable layer that owns them. Do not copy live source-system data into Project context without a clear ownership and refresh model. ## Agent-assisted setup The Agent is part of the builder experience, not only an operator surface. It can help configure its own working environment while the dedicated product surfaces remain the canonical place to review and manage the resulting configuration. Two common examples are: * **Skills** — give the Agent representative work, examples and constraints. It can draft a reusable Skill and propose updates for review. * **Connections** — for supported providers, the Agent can initiate the connection flow. A person still completes the provider authorisation step, and credentials remain outside ordinary Agent context. Review Skills under **Intelligence** and Connections under **Control**. Treat the Agent conversation as the guided setup interface, not as the system of record for configuration. Put instructions and data at the narrowest durable layer that owns them. A company-wide fact should not be copied into every Skill, a rule that belongs only to invoice matching should not become a general Project instruction, and a live ERP record should normally remain in the ERP rather than being copied into Memory. ## Workflows Use a Workflow when a part of the operation has a sufficiently defined path and repeatability matters more than open-ended interpretation. Agent work and Workflows can coexist in the same Project. The Agent can handle variable work and exceptions while Workflows run stable background steps. ## Related documentation * [Instructions and behaviour](/build-and-improve/overview/instructions-and-behaviour.md) * [Skills and operating methods](/build-and-improve/overview/skills-and-operating-methods.md) * [Connect systems](/build-and-improve/overview/connect-systems.md) * [Project context](/build-and-improve/overview/give-the-agent-context.md) * [Test Agent behaviour](/build-and-improve/overview/test-and-improve-the-agent.md) * [Prepare the Project for operators](/build-and-improve/overview/prepare-the-agent-for-operators.md) --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.lleverage.ai/build-and-improve/overview.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.