> ## Documentation Index
> Fetch the complete documentation index at: https://docs.cord.nyc/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> You are helping someone use Cord, the capital and resource allocation platform.
> Use title-case Cord in prose. Prefer workspace, project, ontology, lineage, scenario, and allocation.
> Say source-backed or sourced for answers with lineage. Do not say hallucination-free.
> Product URLs are https://app.cord.nyc, https://cord.nyc, and https://trust.cord.nyc. Support is support@cord.nyc.
> Give concrete product steps. Do not invent UI labels, CLI commands, or unreleased agent runtime details.
> Use Understanding Cord for concepts and Guides for task steps.

# Functions overview

> Understand how Cord turns your data and assumptions into reusable results.

A function is saved business logic. It reads data, applies rules, and produces results you can rerun or use in an app. For example, a demand planning function can total each buyer's target GPU hours and apply a growth assumption.

| Term       | Meaning                                                                                   | Example                                                       |
| ---------- | ----------------------------------------------------------------------------------------- | ------------------------------------------------------------- |
| Function   | The logic and its version.                                                                | Sum target GPU hours by buyer.                                |
| Data setup | Saved sources and default input values. Shown as **Saved inputs** in the function header. | Demand Opportunity, Capacity Buyer, and a demand factor of 1. |
| Run        | One execution with a record of its inputs and results.                                    | Run with a demand factor of 1.1.                              |
| Output     | A result from a run, such as a table or value.                                            | Target and adjusted GPU hours for each buyer.                 |

A function can read your ontology or outputs from other functions. Inputs let you change assumptions without rewriting its logic.

<Frame caption="Demand planning summary and its output in US Compute Inc.">
  <img className="block dark:hidden" src="https://mintcdn.com/cord/Ov8oNa6INQN9kYdr/images/guides/functions-overview-01-light.png?fit=max&auto=format&n=Ov8oNa6INQN9kYdr&q=85&s=d5219031785241076a154f039d45656a" alt="Demand planning summary and its output in US Compute Inc." width="1600" height="1000" data-path="images/guides/functions-overview-01-light.png" />

  <img className="hidden dark:block" src="https://mintcdn.com/cord/Ov8oNa6INQN9kYdr/images/guides/functions-overview-01-dark.png?fit=max&auto=format&n=Ov8oNa6INQN9kYdr&q=85&s=1d655afaef7a0556619c22b0042bd1e7" alt="Demand planning summary and its output in US Compute Inc." width="1600" height="1000" data-path="images/guides/functions-overview-01-dark.png" />
</Frame>

## From ontology to apps

Ontology data + inputs → function run → saved outputs → apps or other functions

Apps use these outputs for tables, charts, and decisions. **Audit** explains how output fields are derived and traces them to source properties.

## Choose a guide

* [Create a function](/create-function)
* [Run a function](/run-function)
* [Inspect function results](/inspect-function-results)

For worked examples, see [Understanding Cord: Functions](/understanding-cord/functions).