# The network

> Where games.sgit.ai sits in the sgit.ai family, and which sibling site answers which question.

*Source: <https://games.sgit.ai/network/index.html> · site v0.5.0 · this file is generated from the same content
as the page, so the two cannot drift. Every page on this site has a `.md` twin; internal links
below point at them.*

---

[Home](../index.md) / The network

# The network

This is one of a family of focused sites on `*.sgit.ai`, each taking one question further than a section could. They share a design, a release discipline, and a habit of publishing the argument before the thing exists.

## The project this site is part of

[**RiskMandate.ai**](https://riskmandate.ai) — *the business risk layer for autonomous systems*. *Agents act. You own the risk.* It begins where security stops: not the finding, but who accepts it, who funds the fix, and who owns the consequence. Its unit is the **mandate** — the right to act, granted by a named owner, scoped, and time-bound rather than standing — and its signature mechanic is that a real risk has [no deny button](https://riskmandate.ai/v0/v0.10/v0.10.0/index.html), only an interval and an owner.

These games are the front end of one part of that: getting a real person to state what they wanted, so the gap against what was granted becomes visible and countable. [How the two connect](../method/grant-vs-mandate.md) · [The grant is not the mandate](https://riskmandate.ai/v0/v0.11/v0.11.0/index.html) · [Library](https://riskmandate.ai/library.html).

## The ones this site leans on

| Site | What it answers | Why it matters here |
|---|---|---|
| [sgit.ai](https://sgit.ai) | the encrypted git for humans and AI agents | The vault layer the games ship in, and the [page on this vault](https://sgit.ai/demos/vaults/agent-permission-games/) with its credential audit |
| [pki.sgit.ai](https://pki.sgit.ai) | a key registry for agents | The mesh, the capability primitives, the nine public profiles and the reductions both games run on |
| [graphs.sgit.ai](https://graphs.sgit.ai) | a node is just a node; meaning is in the edges | The grammar the ideas graph and the mesh are both built under — which is why they join without a mapping |
| [risks.sgit.ai](https://risks.sgit.ai) | you cannot deny a risk, only accept it for a time | Where the delta goes once you have one |
| [llms.sgit.ai](https://llms.sgit.ai) | calling a model without holding an API key | The `sg.llm.*` bridge behind the games' chat panel — the reason we embed the real host |

## And the one this site is for

[**what-can-it-do.games.sgit.ai**](https://what-can-it-do.games.sgit.ai) — the first game, player-facing. It is the family's first fourth-level domain and its first site whose audience is not a practitioner.

The rule that produced it, stated once so the next game inherits it: **a game gets its own domain when it is meant to be shared, and the domain is the game's name so the link explains itself.** If more games reach `measured`, they get `<name>.games.sgit.ai` on the same rule. Everything below that bar lives on a page here.

---

*[Site index for agents](../llms.txt) · [HTML version](https://games.sgit.ai/network/index.html)*