CopilotKit Runtime
@copilotkit/runtime is the server half of CopilotKit: it accepts AG-UI protocol
requests, dispatches them to an AbstractAgent (built-in or external), runs the
stream through an AgentRunner, and responds as Server-Sent Events.
This SKILL.md is the index. Read the reference under references/ that matches
your task — do not try to absorb the whole package from this file.
Mental Model — the three dictionaries you hand to CopilotRuntime
new CopilotRuntime({
agents, // Record<string, AbstractAgent> — see wiring-external-agents or built-in-agent
runner, // AgentRunner (optional) — see agent-runners
intelligence, // CopilotKitIntelligence (optional) — see intelligence-mode (auto-wires runner)
mcpApps, // McpAppsConfig (optional) — see wiring-mcp-apps-middleware
a2ui, // A2UIConfig (optional) — see packages/a2ui-renderer skill
hooks, // { onRequest, onBeforeHandler } — see middleware
beforeRequestMiddleware,
afterRequestMiddleware, // legacy — see middleware
transcription, // TranscriptionService (optional) — see transcription
});
You then mount it:
import { createCopilotRuntimeHandler } from "@copilotkit/runtime/v2";
const handler = createCopilotRuntimeHandler({
runtime,
basePath: "/api/copilotkit",
});
export default { fetch: handler };
When to load which reference
| Task |
Reference |
| Mounting on any fetch-native server (Cloudflare Workers, Bun, Deno, Vercel Edge, Next.js App Router, React Router v7, TanStack Start) or delegating from Express/Node |
references/setup-endpoint.md |
Auth / logging / rate-limit / request-scoped guards via hooks.onRequest / hooks.onBeforeHandler (preferred) or legacy beforeRequestMiddleware / afterRequestMiddleware |
references/middleware.md |
Choosing between InMemoryAgentRunner, SqliteAgentRunner, or a custom subclass — including thread-locking semantics and the runner/Intelligence mutual exclusion |
references/agent-runners.md (+ -in-memory.md, -sqlite.md, -custom.md for backend-specific detail) |
| Enabling durable threads + realtime websocket via CopilotKit Intelligence (a managed service, not self-hostable) |
references/intelligence-mode.md |
Voice transcription — implementing a TranscriptionService subclass for the /transcribe endpoint |
references/transcription.md |
Instantiating BuiltInAgent — Simple Mode (classic) or Factory Mode with TanStack AI (preferred AG-UI-compliant default), AI SDK, or custom factory |
references/built-in-agent.md (+ -factory-modes.md, -helper-utilities.md, -model-identifiers.md) |
Defining server-side tools via defineTool for BuiltInAgent.config.tools (Simple Mode only) |
references/server-side-tools.md |
Wiring an external agent framework into CopilotRuntime({ agents }) |
references/wiring-external-agents.md (index) + per-framework refs (wiring-mastra.md, wiring-langgraph.md, wiring-crewai-crews.md, wiring-crewai-flows.md, wiring-pydantic-ai.md, wiring-adk.md, wiring-llamaindex.md, wiring-agno.md, wiring-aws-strands.md, wiring-ms-agent-framework.md, wiring-ag2.md, wiring-a2a.md) |
| Wiring MCP Apps (runtime-level middleware, not an agent) |
references/wiring-mcp-apps-middleware.md |
Invariants and gotchas (load-once, before any reference)
createCopilotRuntimeHandler is the canonical primitive. createCopilotExpressHandler / createCopilotHonoHandler exist but are avoid at all costs — delegate from Express/Hono routes to the fetch primitive instead.publicLicenseKey is the canonical provider-side field. publicApiKey is a deprecated alias — expect to see it in legacy code, emit the canonical name in new code.- Intelligence mode auto-wires
IntelligenceAgentRunner. Passing both runner and intelligence to CopilotRuntime is rejected at construction.
- Intelligence mode targets the managed CopilotKit Intelligence service (
api.cloud.copilotkit.ai) and is not self-hostable.
hooks.onRequest runs before beforeRequestMiddleware (hook-based middleware wins for Response short-circuits). beforeRequestMiddleware runs after hooks.onRequest (see fetch-handler.ts:136-147).identifyUser (Intelligence) does not forward thrown Response objects — convert to 500. Gate auth rejection in hooks.onRequest, which does forward Responses.agents__unsafe_dev_only and selfManagedAgents are dev-only aliases of each other; do not reach for them in production. Either signals that the SPA is in dev mode.
Reading order for a first-time reader
setup-endpoint — the primitive.built-in-agent or pick one from wiring-external-agents — the agent.agent-runners — production persistence choice.- Optional:
middleware, intelligence-mode, server-side-tools, transcription.
1---2name: runtime3description: @copilotkit/runtime — mount a fetch-native CopilotRuntime on any JS server, wire middleware, pick an AgentRunner, instantiate BuiltInAgent (Factory Mode with TanStack AI is the preferred default) or plug in any of 12 external agent frameworks (Mastra, LangGraph, CrewAI Crews/Flows, PydanticAI, ADK, LlamaIndex, Agno, AWS Strands, MS Agent Framework, AG2, A2A), enable Intelligence mode for durable threads + websocket, register server-side tools via defineTool, and wire voice transcription. Uses the fetch-based createCopilotRuntimeHandler primitive — the Express/Hono adapters are discouraged. Load the reference under references/ that matches your task.4---5
6# CopilotKit Runtime
7
8`@copilotkit/runtime` is the server half of CopilotKit: it accepts AG-UI protocol
9requests, dispatches them to an `AbstractAgent` (built-in or external), runs the
10stream through an `AgentRunner`, and responds as Server-Sent Events.
11
12This SKILL.md is the **index**. Read the reference under `references/` that matches
13your task — do not try to absorb the whole package from this file.
14
15## Mental Model — the three dictionaries you hand to `CopilotRuntime`
16
17```ts
18new CopilotRuntime({
19 agents, // Record<string, AbstractAgent> — see wiring-external-agents or built-in-agent
20 runner, // AgentRunner (optional) — see agent-runners
21 intelligence, // CopilotKitIntelligence (optional) — see intelligence-mode (auto-wires runner)
22 mcpApps, // McpAppsConfig (optional) — see wiring-mcp-apps-middleware
23 a2ui, // A2UIConfig (optional) — see packages/a2ui-renderer skill
24 hooks, // { onRequest, onBeforeHandler } — see middleware
25 beforeRequestMiddleware,
26 afterRequestMiddleware, // legacy — see middleware
27 transcription, // TranscriptionService (optional) — see transcription
28});
29```
30
31You then mount it:
32
33```ts
34import { createCopilotRuntimeHandler } from "@copilotkit/runtime/v2";
35const handler = createCopilotRuntimeHandler({
36 runtime,
37 basePath: "/api/copilotkit",
38});
39export default { fetch: handler };
40```
41
42## When to load which reference
43
44| Task | Reference |
45| ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
46| Mounting on any fetch-native server (Cloudflare Workers, Bun, Deno, Vercel Edge, Next.js App Router, React Router v7, TanStack Start) or delegating from Express/Node | `references/setup-endpoint.md` |
47| Auth / logging / rate-limit / request-scoped guards via `hooks.onRequest` / `hooks.onBeforeHandler` (preferred) or legacy `beforeRequestMiddleware` / `afterRequestMiddleware` | `references/middleware.md` |
48| Choosing between `InMemoryAgentRunner`, `SqliteAgentRunner`, or a custom subclass — including thread-locking semantics and the runner/Intelligence mutual exclusion | `references/agent-runners.md` (+ `-in-memory.md`, `-sqlite.md`, `-custom.md` for backend-specific detail) |
49| Enabling durable threads + realtime websocket via CopilotKit Intelligence (a **managed service**, not self-hostable) | `references/intelligence-mode.md` |
50| Voice transcription — implementing a `TranscriptionService` subclass for the `/transcribe` endpoint | `references/transcription.md` |
51| Instantiating `BuiltInAgent` — Simple Mode (classic) or Factory Mode with TanStack AI (preferred AG-UI-compliant default), AI SDK, or custom factory | `references/built-in-agent.md` (+ `-factory-modes.md`, `-helper-utilities.md`, `-model-identifiers.md`) |
52| Defining server-side tools via `defineTool` for `BuiltInAgent.config.tools` (Simple Mode only) | `references/server-side-tools.md` |
53| Wiring an external agent framework into `CopilotRuntime({ agents })` | `references/wiring-external-agents.md` (index) + per-framework refs (`wiring-mastra.md`, `wiring-langgraph.md`, `wiring-crewai-crews.md`, `wiring-crewai-flows.md`, `wiring-pydantic-ai.md`, `wiring-adk.md`, `wiring-llamaindex.md`, `wiring-agno.md`, `wiring-aws-strands.md`, `wiring-ms-agent-framework.md`, `wiring-ag2.md`, `wiring-a2a.md`) |
54| Wiring MCP Apps (runtime-level middleware, not an agent) | `references/wiring-mcp-apps-middleware.md` |
55
56## Invariants and gotchas (load-once, before any reference)
57
58- `createCopilotRuntimeHandler` is the canonical primitive. `createCopilotExpressHandler` / `createCopilotHonoHandler` exist but are **avoid at all costs** — delegate from Express/Hono routes to the fetch primitive instead.
59- `publicLicenseKey` is the canonical provider-side field. `publicApiKey` is a **deprecated alias** — expect to see it in legacy code, emit the canonical name in new code.
60- Intelligence mode auto-wires `IntelligenceAgentRunner`. Passing both `runner` and `intelligence` to `CopilotRuntime` is rejected at construction.
61- Intelligence mode targets the managed CopilotKit Intelligence service (`api.cloud.copilotkit.ai`) and is **not self-hostable**.
62- `hooks.onRequest` runs **before** `beforeRequestMiddleware` (hook-based middleware wins for Response short-circuits). `beforeRequestMiddleware` runs **after** `hooks.onRequest` (see `fetch-handler.ts:136-147`).
63- `identifyUser` (Intelligence) does **not** forward thrown `Response` objects — convert to 500. Gate auth rejection in `hooks.onRequest`, which does forward Responses.
64- `agents__unsafe_dev_only` and `selfManagedAgents` are dev-only aliases of each other; do not reach for them in production. Either signals that the SPA is in dev mode.
65
66## Reading order for a first-time reader
67
681. `setup-endpoint` — the primitive.
692. `built-in-agent` **or** pick one from `wiring-external-agents` — the agent.
703. `agent-runners` — production persistence choice.
714. Optional: `middleware`, `intelligence-mode`, `server-side-tools`, `transcription`.
om-scogo