> For the complete documentation index, see [llms.txt](https://docs.gage.cash/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.gage.cash/start-here/what-we-are-building.md).

# What we are building

A short view of what gage has shipped, what is in development, and what remains a product vision.

**Status at 18 October 2026.** These are product directions, not launch dates. Only what the app offers today is a live feature; a feature in development or a documented plan is not. Check the app and the current deployment before using any feature with real funds.

## Built on mainnet

gage offers fixed-cost loans against eligible tokens, funded by [partner vaults](/earn/partner-vaults.md) that independent managers run in USDG and in stock tokens, and [indicative suggested costs](/start-here/ratings.md) for specific collateral, amounts and terms. The model uses market history and executable exit evidence to help price long-tail collateral; its suggestions are not guaranteed credit prices. On the same loans:

* [**Leverage**](/borrow/leverage.md)**.** One press loops a USDG loan into more of the token you pledge, in one transaction: one loan and a fixed cost. No liquidation: nothing can touch the collateral before the end date; at the end you choose.
* [**Private borrowing**](/borrow/private.md)**.** Borrow against a screened deposit in a shielded pool, for memes and RH ECO tokens. Deposits, loan terms and positions stay public onchain; the link between a deposit and its loan is not.
* [**Rollover**](/borrow/the-decision.md#roll-over)**.** Repay a running loan and borrow again on the same pledge in one transaction, by hand in its last day or by a standing instruction gage's keeper runs.

The [AI agent tools](/ai-agents/overview.md) are live too: a hosted MCP server, an HTTP API and a TypeScript SDK that read gage and prepare transactions for the wallet's owner to sign, and **Valet**, which lets an agent sign from a wallet of its own under spend limits, by default 100 USDG a loan or a deposit and 250 USDG a day of each.

## In development

* **Agent-run vaults.** Steward is an AI manager built for a partner vault, with a human owner, spending budget and approval boundaries. The broader goal is for anyone to launch a vault and choose an agent to help run it. Agent-run vaults are separate from the [AI agent tools](/ai-agents/overview.md), with which your own agent prepares and you sign.

## Product vision; not built or offered yet

* **Team Runway credit lines.** A separate fixed-cost credit line would let token teams seek funding against their own tokens, with dedicated eligibility and underwriting rules. This is a design concept, not an available borrowing route.
* **More complete privacy.** Beyond the private lane, we want to explore private borrower positions and private vault balances or strategies.
* **Long-tail pricing data as a product.** gage wants to package and sell its proprietary, model-derived long-tail collateral and fixed-term loan pricing signals as a data provider to LLM and agent platforms, lenders and other integrators. The first-party suggested-cost model exists; a paid external feed, provider integration and commercial terms do not.


---

# 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.gage.cash/start-here/what-we-are-building.md?ask=<question>&goal=<user_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.