# AISquare > AISquare turns agent runs into governed, explainable decisions. Govern any agent, then run the Trust Loop on every decision (Understand, Prevent, Fix, Remember) so that Collective Intelligence compounds and your teams ship it as Studios. - AISquare is a governance and observability layer that wraps any agent framework, model, or workflow. It is not a model provider or an agent framework itself. - The narrative spine is Govern, then Understand, Prevent, Fix, and Remember, then Collective Intelligence, then Studios. - Every page is available as Markdown: append `.md` to any docs URL, or send `Accept: text/markdown`. The full corpus is at /llms-full.txt. - Questions: bots@aisquare.studio. ## Getting Started - [Overview](https://docs.aisquare.studio/docs/getting-started/overview): Govern your agents, run the Trust Loop on every decision, compound the result as Collective Intelligence, and ship it as Studios. - [Quickstart](https://docs.aisquare.studio/docs/getting-started/quickstart): Make your first governed call against the AISquare API, complete and copy-paste runnable, then read the response. - [Authentication](https://docs.aisquare.studio/docs/getting-started/authentication): Create a workspace API key, store it as an environment variable, and authenticate every request as the right user. - [Core Concepts](https://docs.aisquare.studio/docs/getting-started/core-concepts): The entities behind the AISquare API and how they fit together, so the responses make sense. ## Govern - [Connect your agents](https://docs.aisquare.studio/docs/govern/connect-your-agents): Two ways to send your agents to AISquare, the SDK and the multi-provider proxy. Both give full observability, governance, and runtime enforcement. They differ on where the gate runs. - [Agent identity](https://docs.aisquare.studio/docs/govern/agent-identity): Every trace routes by agent name. How identity is set on each integration path, how to run more than one agent, and what failure looks like when a name is missing. - [Proxy integration](https://docs.aisquare.studio/docs/govern/proxy-integration): Put your AI agents behind the AISquare proxy. A drop-in gateway that records every run and enforces your Rule Book, for Claude, OpenAI, Azure OpenAI, and Gemini, plus the Claude Code and Codex CLIs. - [Understand](https://docs.aisquare.studio/docs/govern/understand): See why an agent decided what it did, on the reasoning graph, the claims it made, the evidence behind them, and the assumptions it filled in. - [Prevent](https://docs.aisquare.studio/docs/govern/prevent): Decision-level policy gates. Evaluate every run against your rule book, and block, improve, or warn on an action before it runs. - [Fix](https://docs.aisquare.studio/docs/govern/fix): The correction loop. Fix a wrong or blocked decision in context, on the reasoning that produced it, and carry the correction forward. - [Remember](https://docs.aisquare.studio/docs/govern/remember): Enterprise memory. Every decision, intervention, and correction is persisted as reusable context the next run can draw on. ## Studios - [Studios Overview](https://docs.aisquare.studio/docs/studios/overview): A Studio turns governed collective intelligence into experiences your customers ship: a reusable unit of knowledge, AI, and humans that draws on the reasoning and memory graph. - [Experiences and resources](https://docs.aisquare.studio/docs/studios/experiences-and-resources): The content model inside a Studio. An experience is a container; a resource is the item inside it. Resources come in four types: AI Expert, Note, Quest, and Podcast. - [Collections](https://docs.aisquare.studio/docs/studios/collections): A collection is a curated, ordered group of experiences inside a Studio. Use collections to build learning paths and courses, and to control who can see them. ## Guides - [Integration patterns](https://docs.aisquare.studio/docs/guides/integration-patterns): How AISquare fits into different application architectures, and which pattern to pick. - [User onboarding](https://docs.aisquare.studio/docs/guides/user-onboarding): How users are created, authenticated, and managed when you build on AISquare. - [Token lifecycle](https://docs.aisquare.studio/docs/guides/token-lifecycle): How access and refresh tokens work in AISquare and how to manage them in production. - [Content modeling](https://docs.aisquare.studio/docs/guides/content-modeling): How AISquare structures content and how to map it onto your UI. - [Search and filtering](https://docs.aisquare.studio/docs/guides/search-and-filtering): Retrieve and refine data efficiently with search, filters, sorting, and pagination. - [Permissions](https://docs.aisquare.studio/docs/guides/permissions): How access to AISquare content is controlled and how to handle permissions correctly. ## Agent Orchestrator - [Overview](https://docs.aisquare.studio/docs/agent-orchestrator/overview): Run several Claude Code sessions on one problem, coordinated through a shared working memory. One planner you talk to, coders that pick work up, a runner that verifies. - [Quickstart](https://docs.aisquare.studio/docs/agent-orchestrator/quickstart): One Claude account, five terminals, a working AI team in two minutes. - [The board](https://docs.aisquare.studio/docs/agent-orchestrator/board): The live cockpit for your AI team. Session states, the task table, the event feed, a detail bar, and a jump into any session's transcript. - [Tasks and the work cycle](https://docs.aisquare.studio/docs/agent-orchestrator/tasks): How work flows through the team: idempotent adds, atomic claims, dependencies, review, and reopen feedback that finds its way back. - [Remote agents (MCP)](https://docs.aisquare.studio/docs/agent-orchestrator/remote-agents): Let Claude clients outside your terminals join the board: the desktop app, a browser-testing agent, another machine. - [Memory and reference](https://docs.aisquare.studio/docs/agent-orchestrator/memory-and-reference): The long-term brain (gbrain distiller and recall), every environment knob, and the troubleshooting list. ## API Reference - [API Reference](https://docs.aisquare.studio/docs/api-reference/overview): Generated, per-operation API reference with a Try-it playground. ## Reference - [Compliance and audit](https://docs.aisquare.studio/docs/reference/compliance): How the signed, per-decision record maps to what the EU AI Act, ISO/IEC 42001, and the NIST AI RMF ask you to produce. - [Errors](https://docs.aisquare.studio/docs/reference/errors): How AISquare APIs return errors and how to handle them. - [Support](https://docs.aisquare.studio/docs/reference/support): Get help when working with AISquare APIs. Contact bots@aisquare.studio. ## More - [Collective Intelligence](https://docs.aisquare.studio/docs/collective-intelligence): The reasoning-aware, auditable, governable layer with the memory and context of your agents and your humans. Everything writes into it; everything is powered by it. - [Case study: Dramatyze](https://docs.aisquare.studio/docs/case-study): Built in about 30 hours of focused work: how the Dramatyze team shipped a customer-facing product while dogfooding AISquare. - [Changelog](https://docs.aisquare.studio/docs/changelog): API and product changes for AISquare, with RSS and Markdown feeds. - [Use these docs in your agent](https://docs.aisquare.studio/docs/use-in-your-agent): Query these docs as a tool over MCP, pull the whole corpus as plain text, or copy any page as Markdown. ## Optional - [AISquare website](https://aisquare.studio/): The product and company. - [GitHub organization](https://github.com/AISquare-Studio): The open source AISquare ecosystem. - [Support](mailto:bots@aisquare.studio): Reach the team.