Open Source · Pre-GA · Apache 2.0
Build, host, and
orchestrate tool-using
AI agents.
Agentrail gives you a composable runtime, a hosted server layer, multi-agent orchestration, and a clear extension model — without locking you into any platform.
import { defineAgent } from "@agentrail/runtime-core"; import { createWebSearchTool } from "@agentrail/tools"; export const researchAgent = defineAgent({ name: "researcher", model: { provider: "anthropic", modelId: "claude-opus-4-5", apiKey: process.env.ANTHROPIC_API_KEY, }, systemPrompt: "You are a research assistant.", tools: [createWebSearchTool()], }); // stream a response const result = await researchAgent.stream( "What are the latest advances in agentic AI?", );
Why Agentrail
The right level of abstraction.
Most teams land between two extremes — scripts that break at scale and platforms that make you dependent. Agentrail is designed to sit in between.
Too little structure
Ad hoc scripts
- No session management
- Tool contracts are implicit
- Orchestration is bespoke every time
- Testing is painful
- Hard to extend without rewriting
Just right
Agentrail
- Composable runtime with clear contracts
- Host layer for chat & stream lifecycles
- First-class multi-agent orchestration
- Plugin & context provider extension model
- No platform lock-in, fully self-hosted
Too much lock-in
Hosted platforms
- Opaque internals
- Metered billing surprises
- Limited extensibility
- Vendor controls your data
- GUI-first, not code-first
What's included
Everything you need to ship agents.
A focused set of packages — each with a clear responsibility, each independently usable.
@agentrail/runtime-core
Agent Runtime
Define agents with typed tool contracts. The LLM loop, tool dispatch, provider abstractions, and usage tracking — all in one stable core.
@agentrail/host
Host Layer
Chat and stream request lifecycles, profile resolution, session context assembly, and a plugin system for cross-cutting behavior.
@agentrail/orchestration
Multi-Agent Orchestration
Spawn sub-agents, send typed work, wait on conditions, and recover from failures. Persistent state across restarts via JSONL-backed mailboxes.
@agentrail/prompts
Prompt SDK
Compose system prompts from versioned fragments and bundles. File-based prompt authoring with hot-reload and variable interpolation.
@agentrail/memo
Session Memory
Filesystem-backed session storage with append-only JSONL history, automatic context compaction, and conversation branching.
@agentrail/sandbox
Sandboxed Execution
Docker-based isolated execution environment. Browser automation, shell commands, and file I/O — all safely contained per session.
Architecture
Layered by design.
Start with the opinionated SDK, then drop to lower-level primitives only when you need tighter control.