$kage

zed ​

kage rpc is a spec-conformant Agent Client Protocol agent: JSON-RPC 2.0 over stdio, one message per line (newline-delimited), protocol version 1. Zed speaks ACP natively, so it drives kage the same way it drives its other external agents - no shim, no wrapper.

start the agent ​

sh
kage rpc

Optional flags:

  • -m, --model <provider:model> pins the model for the connection.
  • --system <text> overrides the system-prompt role.

Credentials resolve as for the TUI and print mode (OS keyring, kage auth login, or an API-key env var). With no provider configured kage rpc prints a message and exits non-zero.

configure zed ​

Add kage as an agent server in Zed's settings.json:

json
{
  "agent_servers": {
    "kage": {
      "command": "kage",
      "args": ["rpc"]
    }
  }
}

Use an absolute command if kage is not on Zed's PATH; pass a model with "args": ["rpc", "-m", "anthropic:claude-sonnet-4-6"]. Pick kage from Zed's agent panel and prompt as usual. Tool calls surface as Zed permission prompts (kage never auto-approves); the agent's text and reasoning stream in as it works.

what kage implements ​

Client to agent:

methodparams -> result
initialize{protocolVersion, clientCapabilities} -> {protocolVersion, agentCapabilities, agentInfo, authMethods}
session/new{cwd, mcpServers} -> {sessionId}
session/load{sessionId, cwd, mcpServers} -> replays history as session/update, then null
session/prompt{sessionId, prompt: ContentBlock[]} -> {stopReason}
session/cancelnotification {sessionId}

Agent to client:

  • session/update notification {sessionId, update} where update is tagged by sessionUpdate: agent_message_chunk, agent_thought_chunk, tool_call, tool_call_update.
  • session/request_permission request {sessionId, toolCall, options} -> {outcome}; the client returns {outcome: {outcome: "selected", optionId}} or {outcome: {outcome: "cancelled"}}. kage blocks the tool until the client answers and never auto-approves.

kage advertises agentCapabilities.loadSession: true, empty authMethods, and agentInfo {name: "kage", version}. It does not request fs/terminal client capabilities: kage runs its own tools in-process and gates them through session/request_permission.

hand-driven smoke test ​

sh
printf '%s\n%s\n' \
  '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":1,"clientCapabilities":{}}}' \
  '{"jsonrpc":"2.0","id":2,"method":"session/new","params":{"cwd":"/tmp","mcpServers":[]}}' \
  | kage rpc

Two framed JSON-RPC results come back, one per line: the agent's capabilities, then a fresh sessionId.