> For the complete documentation index, see [llms.txt](https://docs.trueopen.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.trueopen.ai/protocol/trust-model.md). # Trust Model Authority, verification boundaries, and security assumptions in TrueOpen. TrueOpen distributes execution, data delivery, verification, and consensus across different roles. This page explains their boundaries. The [V1 protocol](/protocol/v1.md) defines the corresponding acceptance and failure rules. ## Chain and consensus Accepted chain state is the authority for assignments, deadlines, balances, and finality. This relies on the chain's consensus security assumptions and correct deterministic execution by validators. An event or indexer response is an observation that should be reconciled with accepted state. ## Builders and data availability Builders coordinate proposals and transport Task material through Nexus. Commitments let recipients check that delivered bytes match the accepted Task context. Commitments alone do not guarantee that data remains available: retention, availability duties, and defined failure paths are also required. A Builder acknowledgement establishes transport progress, not a final verdict or payment. The [data and evidence rules](/protocol/v1/data-and-evidence.md) define the relevant obligations. ## Workers, Verifiers, and Profiles Workers provide inference results and evidence. Verifiers independently evaluate the material under the selected Profile. Selection, role separation, commit-reveal, and challenges limit specific manipulation opportunities; their protection depends on the verification algorithm and the participating set. A registered Profile fixes the model, execution, verification, and evidence context. Registration does not itself prove universal model quality or factual correctness. ## Cortex and model runtimes Cortex controls provider signing and validates the bindings returned by the model adapter. The runtime must not hold the service key or decide chain outcomes. Operators remain responsible for protecting their keys and host. See [Cortex architecture](/providers/model-integration/cortex-architecture.md). ## Governance and assets Governance operates within versioned constraints and cannot rewrite historical Tasks through an ordinary parameter change. The canonical asset route adds bridge-signature and origin-chain assumptions beyond local chain consensus. A local withdrawal burn and delivery on the origin chain are separate outcomes. See [Governance](/protocol/v1/governance.md) and [Cross-chain assets](/protocol/v1/cross-chain-assets.md) for these boundaries and their failure controls. --- # 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 dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.trueopen.ai/protocol/trust-model.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. 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.