This documentation is the entry point for the open-source `seshat` runtime: CLI, Go SDK, core concepts, and the platform layers built on top of it. Use this page to choose the right path before diving into articles.
The open-source Go runtime: agent loop, tools, permissions, MCP, RAG, sessions, CLI, gRPC, and SDK.
The product platform built on top: multi-user API, organizations, workspaces, audit, scheduler, desktop application, and governance.
curl -fsSL https://raw.githubusercontent.com/EngineerProjects/seshat/main/scripts/install.sh | bashPick the entry point that matches what you are trying to do. This page is meant to orient, not to act like a landing page.
Install the binary, configure a provider, and run your first autonomous agent locally.
Use the Go runtime inside your own application once the SDK reference lands.
Learn how the runtime, tools, MCP, memory, and platform layers fit together.
Navigate from the runtime docs toward SeshatOS, the multi-user API and desktop layer built on top.
Understand the project vision, the two repositories, and where to contribute across the runtime, CLI app, desktop app, and future SDKs.
CLI, Go SDK, gRPC clients, and later platform UIs consume the same runtime foundation.
`seshat` provides tools, permissions, memory, MCP, RAG, sessions, execution, and agent primitives.
`seshat-ai` / SeshatOS adds multi-user APIs, workspaces, quotas, audit, scheduled jobs, and desktop UX.
The structure is broader than the currently published content. This page reflects the real state instead of pretending every section is already complete.
Install Seshat and run your first autonomous agent in minutes.
Understand the architecture and core ideas that make Seshat different.
Programmatic access to Seshat from any language or platform.
Step-by-step guides for end users and product surfaces.
Deep dives into agentic AI concepts, patterns and architecture decisions.
Honest side-by-side comparisons with popular agent frameworks.