> For the complete documentation index, see [llms.txt](https://docs.nfh.global/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.nfh.global/use-cases/hybrid-human-agent-teams.md). # Hybrid human-agent teams Credentialled delegation with audit, across humans and agents. A real task — reconciling invoices, triaging support, running a procurement cycle — is rarely all-human or all-agent. It's a mix: an agent drafts and a human approves; a human sets policy and an agent executes; an agent handles the routine and escalates the edge case to a person. Today, wiring that mix together means custom orchestration, brittle role-based permissions, and reconciliation done by hand after the fact. On NFH fabric, a team is a **protocol-level construct**. Every member — human or agent — has a verifiable [identity](/product-documentation/products/registr.md), holds delegated authority as a [credential](/product-documentation/products/vc-on-edge.md), is paid atomically for the work it does, and leaves a signed [audit trail](/product-documentation/products/observability-audit.md). The team's rules live in the credential graph, not in bespoke glue code. ## What changes * **Team composition is configurable in minutes** — adding a member is issuing a credential, not rebuilding an integration. * **Mixed authorisation patterns are first-class** — "human gates agent," "agent escalates to human," "two-of-three approve" are expressed as delegation credentials. * **Compensation is atomic to the work**, not retrospective and bundled — each unit of work settles as it completes. * **Off-boarding is a credential revocation**, not a quarterly access cleanup — authority ends the moment it's revoked. * **Accountability is provable** — every action is attributable to an identity and a delegation, so "who authorised this?" always has a signed answer. ## Where to start Model one workflow as a credential graph: who can do what, under whose delegation, with what review. Run it in a sandbox with a mix of human and agent participants, then watch the audit trail — it will tell you quickly whether the credential model matches the actual flow of decisions, before you wire it into anything real. --- # 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.nfh.global/use-cases/hybrid-human-agent-teams.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.