Updates `@agentclientprotocol/sdk` to 1.9.0 (schema v1.26.0, v2.0.0-alpha.9) and adapts the agent to the v2 spec clarifications in that release. ## Changes **Client-provided system prompt (unstable)** - Advertises `sessionCapabilities.systemPrompt` with `mode.override` on v1 and `session.systemPrompt` on v2. - Accepted on `session/new`, `load`, `resume`, and `fork`: - `append` adds the text after the Claude Code preset. - `override` replaces the preset with the text. Tools, MCP servers, permissions, and settings (including CLAUDE.md through `settingSources`) are unchanged. - An unsupported mode or non-text content fails with `invalid_params` before any session is created, reused, or forked. - Takes precedence over the `_meta.systemPrompt` extension. AIR custom instructions still follow the client text. - A different prompt on resume recreates a live session. Claude Code reuses its saved system prompt until the next compaction, which the RFD allows. **v2 session/resume state snapshot** - Sends the current foreground state (`idle`, `running`, or `requires_action`) after any replay and before the response, as the clarified spec requires. This applies even without a replay. **v2 replay of subagents** - No longer replays a child's past `running`/`idle` as current activity. Each replayed child gets one `state_update` with its last state after the replay. **v2 tool call `_meta` patches** - Reports carry only the top-level keys that changed. `claudeCode` is still sent whole, since clients don't merge nested objects. - A removed key is sent as `null`. - A tool call id that starts again removes the old keys. - A replayed tool call this connection has not reported sends `_meta: null` first, then the current keys. **Cancellation and close** - A `session/close` whose turns don't end within the teardown timeout settles them as cancelled. The client gets `idle` with `cancelled`, or `-32800` for a prompt never inserted, before the close response. - A prompt racing `session/close` ends as cancelled (`-32800` on v2) instead of `-32603`. On v1 it now returns `stopReason: cancelled`, matching a prompt queued just before the close. - A stream error after a cancel ends the work as `cancelled` instead of `error`. ## Not in this PR - After a URL-mode MCP elicitation is accepted, the session reports `running` right away. The spec asks for `requires_action` until the external interaction finishes, but a server may never send `elicitation_complete`, so this needs its own design. - `requires_action` is reported while one foreground subagent waits, even if other foreground work can continue. This is SHOULD-level in the spec. ## Testing - `npm run build`, `npm run check`, and `npm run test:run` pass. - Golden files changed because of the new capability and the resume snapshot. `compare.ts` allows the new capability against origin/main. - `session index cost > reads only the changed transcript for a subscription` failed intermittently locally (about 3 in 15 runs) on a loaded machine. This PR doesn't touch the session index, but watch it in CI. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
ACP adapter for the Claude Agent SDK
Use Claude Agent SDK from ACP-compatible clients!
This tool implements an ACP agent by using the official Claude Agent SDK, supporting:
- Context @-mentions
- Images
- Tool calls (with permission requests)
- Compact file changes through the negotiated AIR diff patch extension
- Following
- Edit review
- TODO lists
- Nested subagent transcripts
- Interactive (and background) terminals
- Custom Slash commands
- Client MCP servers
/mcpin the chat: the MCP server status as a list. The adapter runs/mcp reconnect,/mcp enable, and/mcp disablethrough the SDK control API, because Claude Code refuses them in SDK mode. A reconnect of an ACP server that needs authentication starts MCP OAuth through URL elicitation- Session-scoped long-running goals for AIR through the goal extension under
_meta.jetbrains.air.goal - Structured errors, recovery, and warnings through the opt-in session failure extension
- Concrete model and effort defaults through the opt-in recommended config value extension
- Tool permission presentation, editable choices, and durable effects through the permission extension
- One fact per field in every tool call report, as the ACP tool call contract defines
- All AIR extensions, capabilities, and
_metakeys: AIR extensions
Learn more about the Agent Client Protocol.
To try changes that have landed on main but are not released yet, install from the
preview channel — every push to main publishes one. See
docs/RELEASES.md.
npm install @agentclientprotocol/claude-agent-acp@preview
Subagent sessions
A client that declares clientCapabilities.subagents, other than JetBrains AIR, gets each Agent or
Task subagent as its own child session, following the subagents RFD:
subagent_updateon the parent announces the child before any of its traffic, with the parent'stitleanddescriptionfor it, and reports its work state:running,requires_actionwhile one of its requests is open (andrunningagain when the last one ends, unless a cancel withdrew it: the child is stopping then), andidlewhen its work ends, withend_turnorcancelledwhen the SDK gives the reason. A failure isidlewith theerrorstop reason and the SDK's error text as itserror. The child keeps its session for every delegation to it.- The prompt of the Agent or Task call that launches the child, and the message of each
SendMessage call that resumes it, is a
session_messagein the child's transcript. The tool call stays a tool call of the session that made it. session/cancelon the parent interrupts Claude Code, which stops the foreground children that the turn waits on. Background children keep running, as the RFD prefers; a client stops one with its owncancel. Each stopped child reportsidlewithcancelledfrom Claude Code's own events, and its open requests are withdrawn. The adapter ends no child at a cancel itself, so a child that keeps running keeps its open requests and its open tool calls. To spare background children, the adapter declares the SDK optionperTaskStopAffordance, which a client can turn off in_meta.claudeCode.options. That option spares background workflows too, which a client cannot stop, so the adapter stops them after the interrupt, as Claude Code did without it. Background commands keep running after a cancel either way. A tool call of a stopped child that Claude Code reports no result for, such as the command it was running, ends in the child's session before the cancelled prompt resolves. When Claude Code stops a background child after that, or while no prompt runs, the tool call ends right after the child reportsidle. A child that still runs when Claude Code exits isidlewith no stop reason.- A child whose parent is not yet known when it ends is not exposed, since the RFD forbids guessing its parent.
- A replay restores each child under its live session id, from its own transcript: its prompt and
each SendMessage message are
session_messages, and its end is anidlebefore the response. A launch without Claude Code's metadata file for the child stays an ordinary tool call of the parent, and a notice, for a client that takes notices, says that subagent history is missing. - After a load or resume, with or without replay, a SendMessage that resumes an agent from before it runs that agent in its session from before. Before the agent's traffic, each of its ancestors that the client was not told about in this connection is announced, parent before child and with no work state, as the RFD requires. An agent whose launch the session's history does not record stays unexposed.
- Each running child is announced with the
cancelcapability.session/cancelwith the child's session id stops that child's current work, through the SDK'sQuery.stopTask, and not its parent's: Claude Code also stops the foreground children it waits on, and the parent's turn goes on. Each stopped child reportsidlewithcancelledfrom Claude Code's own events, and its requests that the stop withdraws report norunningfirst. A replayed child, which no longer runs, gets no capabilities. A client cannot prompt or otherwise change a child.
An ACP v2 client gets the same without declaring anything, since subagents
are part of v2's baseline. Each child also reports its work state as its own state_update, right
after the parent's copy.
JetBrains AIR keeps the earlier draft of the RFD (subagent_spawned, subagent_state_update),
which it enables with nativeSubagentSessions in _meta.jetbrains.air.capabilities. See
AIR extensions.
Without either signal, Agent and Task stay ordinary tool calls and child interactions stay on the
root session. Clients that use the historical _meta["subagent-transcript"] capability or the
forwardSubagentText session option retain the flattened child transcript.
Experimental ACP v2
Started with CLAUDE_AGENT_ACP_EXPERIMENTAL_V2=1, the adapter also serves the
draft ACP v2: a client that asks for
protocol version 2 or later in initialize gets v2, and every other client gets v1, as before.
The draft can still change, and this surface changes with it.
On v2 the adapter serves the session baseline (session/new, session/list, session/resume
with replay, session/close, session/prompt, session/cancel), session/delete,
session/fork, additional directories, config options (the mode is one), plans, notices,
compaction, display terminals for commands, permission requests, elicitation, providers, auth, and
subagents. Steering and the AIR extensions are v1 only for now.
Contribution Policy
This project does not require a Contributor License Agreement (CLA). Instead, contributions are accepted under the following terms:
By contributing to this project, you agree that your contributions will be licensed under the Apache License, Version 2.0. You affirm that you have the legal right to submit your work, that you are not including code you do not have rights to, and that you understand contributions are made without requiring a Contributor License Agreement (CLA).