> For the complete documentation index, see [llms.txt](https://docs.suno.finance/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.suno.finance/financial-model/governance-and-auditability.md). # Governance and auditability The engine is a pure function: the same inputs produce the same valuation, with no hidden state, randomness, or manual adjustment. Everything in this page builds on that property, because it makes a strong guarantee possible: any published valuation can be recomputed exactly, from its own stored inputs, by anyone, at any time. #### Published versions A valuation becomes official by being published as a numbered version. Publication freezes: * the complete input snapshot: every number the engine consumed, including project data, telemetry series, macro projections, price curve points, risk scores, scalar assumptions, and the equipment event schedule; * the full year-by-year table, intermediate columns included; * the sensitivity tornado as calibrated at that moment, each row carrying the perturbation magnitudes it applied; * any engine warnings active at publication; * the publisher's identity and timestamp. Published versions are never edited or deleted. Corrections happen by publishing a new version. #### Self-describing snapshots Snapshots carry a schema version, incremented with every change to the snapshot format and documented against its predecessor. Because sensitivity rows embed their own applied magnitudes, annual recalibrations do not orphan history: a version published under an earlier calibration still states exactly what it used. Interpreting an old valuation takes the snapshot and this documentation, nothing else. The engine's arithmetic is protected the same way. A golden-master regression pins a complete reference valuation, every intermediate column included, inside a suite of over 1,500 automated tests. An unintended change to the arithmetic fails the build; an intended methodology change re-pins the golden master with the value delta recorded in the fixture's changelog. The result is a running, quantified history of how the model's output has evolved and why, which is the concrete mechanism behind the rule that methodology improves over time while published valuations stay put. #### The active version and rollback Exactly one published version per project is active at a time; it is the one token holders and the Reserve consume. Publication normally activates the new version, but a version can be staged without activating it, and the pointer can be rolled back to a previous version if a problem is found. Every pointer movement lands in an append-only activation log with author, timestamp, and reason. The consumed NAV changes only by moving a pointer between immutable objects. #### Drafts and diffs Before publication, the draft valuation recomputes live against current data, and the interface diffs the draft's inputs against the active version: which assumptions changed, which telemetry moved. A publisher sees why the number changed before freezing it. Engine warnings surface in the draft and freeze with the version, so data-quality caveats stay attached to the number they qualify. #### Recalibration The empirical calibrations behind the [sensitivity analysis](/financial-model/sensitivity-analysis.md), production variability from portfolio telemetry and FX quantiles from the official exchange-rate series, are refreshed annually with computation dates and sources on record. As operating history grows, backtesting of modeled against realized generation and OPEX, per project and per year, will replace the interim benchmark ranges with the protocol's own measured forecast errors. --- # 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.suno.finance/financial-model/governance-and-auditability.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.