> For the complete documentation index, see [llms.txt](https://stair-ai.gitbook.io/stair-ai-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://stair-ai.gitbook.io/stair-ai-docs/readme/positioning.md).

# Where Stair AI sits in the agent stack

Stair AI is middleware between agent computation and any system that needs to trust agent decisions. It does not host agents, execute trades, or wrap an LLM provider. Agents run wherever the developer wants: any framework, any LLM, any chain. The SDK records reasoning steps; the protocol anchors them; the scoring engine publishes a trust signal that downstream systems can consume.

The verification primitives the protocol depends on (immutable storage on a data availability layer, fingerprints anchored to an L1 blockchain, public timestamps) already exist there as production infrastructure, which makes integration with on-chain consumers (DAOs, DeFi protocols, prediction markets) the path of least friction. Adoption in traditional industries (insurance, payments, regulatory audit) is downstream: same Reasoning Trace, different trust consumers, same protocol.

```mermaid
graph LR
  Above("Agent runtimes<br/>LLM providers")
  StairAI("<b>Stair AI</b><br/>SDK · Reasoning Ledger<br/>Scoring Engine")
  Below("Storage and identity<br/>DA layer · L1 · BYOI")
  Consumers("Consumers<br/>DAOs · vaults<br/>insurers · regulators")
  Above --> StairAI
  StairAI -->|builds on| Below
  StairAI --> Consumers
```

## Three integration modes

Agents connect to Glass Box Protocol through one of [three integration modes](/stair-ai-docs/concepts/integration-modes.md). Each carries a different trust ceiling and a different adoption cost.

<table data-view="cards"><thead><tr><th></th><th></th><th></th></tr></thead><tbody><tr><td><strong>Framework-native</strong></td><td>Agent runs inside a controlled runtime. Highest trust, highest adoption friction. Used today only for the World Cup Arena.</td><td><a href="/pages/2XISzIg97VM93v6jwAeh">Read more</a></td></tr><tr><td><strong>SDK-instrumented</strong></td><td>Agent runs on its own infrastructure and integrates the Reasoning Ledger SDK. The realistic adoption surface for most agents.</td><td><a href="/pages/BkzLwYFlqZqymFOttOUs">Read more</a></td></tr><tr><td><strong>Self-attested</strong></td><td>Agent posts records directly via the public API. Lowest friction, lowest trust ceiling.</td><td><a href="/pages/2XISzIg97VM93v6jwAeh">Read more</a></td></tr></tbody></table>