> 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/understand-lleverage/readme/from-manual-to-autonomous.md). # Automation levels A practical way to think about increasing automation without assuming every process should become fully autonomous. Lleverage can support operational work at different levels of automation. The four levels describe how responsibility is divided between people and Lleverage. A process does not need to progress through every level. ![Manual, Assisted, Supervised and Autonomous as increasing responsibility Lleverage can take. Background Workflow responsibility rises while human involvement shifts toward configured approval and exception boundaries; the right level depends on the process and full autonomy is not a required destination.](https://3370306531-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FlgSglJDQZxDZdn20CUcl%2Fuploads%2Fgit-blob-ee5823e21b13a73e8e22e6b1e09de55751231bf2%2Fdiagram-8-autonomy-ladder.png?alt=media) ## Manual A person performs the work directly in the existing systems. Lleverage may not be involved yet, or may only help make the work and its bottlenecks more visible. ## Assisted A person still owns the work, but the Agent or a Skill helps with part of it. Examples include reading incoming documents, preparing a draft, comparing information across systems, or suggesting the next action while the person remains in control of the result. This is often where the Agent is most visible: the operator is still in the foreground, but less of the work is manual. ## Supervised More of the execution moves into the background, usually through a configured Workflow or other repeatable automation. The system performs the routine path and deliberately calls a person in where judgement, approval or missing information is required. Those handoffs can appear as Requests. The person is no longer driving every step, but they are still part of the designed execution path. ## Autonomous The routine path runs without a person actively driving it. People are mainly involved when an exception, uncertainty or configured decision boundary requires attention. Autonomous does not mean unconstrained. The execution still depends on configured access, instructions, validation, monitoring and human boundaries. ## Selection criteria The appropriate level depends on the process. Consider: * how repeatable the work is; * how costly a wrong action would be; * whether the required information is available and trustworthy; * how easily the result can be verified; * which decisions genuinely require a person. Some work may remain supervised permanently. Other work can move further into the background as the organisation gains evidence that the routine path is dependable. --- # 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/understand-lleverage/readme/from-manual-to-autonomous.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.