chore(main): release 0.25.2 (#511 )

🤖 I have created a release *beep* *boop* --- ## [0.25.2](https://github.com/agentclientprotocol/claude-agent-acp/compare/v0.25.1...v0.25.2) (2026-04-06) ### Bug Fixes * prioritize ANTHROPIC_MODEL env var over settings.model in model … ([#505](https://github.com/agentclientprotocol/claude-agent-acp/issues/505)) ([bea1a40](bea1a40e9b)) --- This PR was generated with [Release Please](https://github.com/googleapis/release-please). See [documentation](https://github.com/googleapis/release-please#release-please). Co-authored-by: acp-release-bot[bot] <246668977+acp-release-bot[bot]@users.noreply.github.com>
fix: prioritize ANTHROPIC_MODEL env var over settings.model in model … (#505 )
2026-10-27 12:40:32 +00:00 · 2026-04-06 21:50:32 +02:00 · 2026-04-06 19:48:24 +00:00 · 2026-04-06 19:31:14 +02:00 · 2026-04-06 09:01:18 +02:00 · 2026-04-06 06:59:16 +00:00
15 changed files with 2155 additions and 482 deletions
--- a/.github/dependabot.yml
+++ b/.github/dependabot.yml
@@ -0,0 +1,18 @@
+version: 2
+updates:
+  - package-ecosystem: "github-actions"
+    directory: "/"
+    schedule:
+      # Check for updates to GitHub Actions every week
+      interval: "weekly"
+  - package-ecosystem: "npm"
+    directory: "/"
+    schedule:
+      interval: "weekly"
+    allow:
+      - dependency-type: "all"
+    groups:
+      minor:
+        update-types:
+          - "minor"
+          - "patch"
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -21,12 +21,12 @@ jobs:
    timeout-minutes: 5

    steps:
-      - uses: actions/checkout@08c6903cd8c0fde910a37f88322edcfb5dd907a8
+      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd

      - name: Use Node.js
-        uses: actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020
+        uses: actions/setup-node@53b83947a5a98c8d113130e565377fae1a50d02f
        with:
-          node-version: latest
+          node-version: "lts/*"
          cache: "npm"

      - name: Install dependencies
--- a/.github/workflows/publish.yml
+++ b/.github/workflows/publish.yml
@@ -1,14 +1,44 @@
 name: Publish and Release
 on:
  push:
-    tags:
-      - "v*.*.*"
-  workflow_dispatch:
+    branches:
+      - main

 jobs:
+  release-please:
+    runs-on: ubuntu-latest
+    environment: release # Optional: for enhanced security
+    permissions:
+      contents: write
+      issues: write
+      pull-requests: write
+    steps:
+      # Generating a GitHub token, so that PRs and tags created by
+      # the release-please-action can trigger actions workflows.
+      - name: Generate GitHub token
+        uses: actions/create-github-app-token@f8d387b68d61c58ab83c6c016672934102569859
+        id: generate-token
+        with:
+          # GitHub App ID secret name
+          app-id: ${{ secrets.RELEASE_PLZ_APP_ID }}
+          # GitHub App private key secret name
+          private-key: ${{ secrets.RELEASE_PLZ_APP_PRIVATE_KEY }}
+      - uses: googleapis/release-please-action@16a9c90856f42705d54a6fda1823352bdc62cf38
+        id: release
+        with:
+          token: ${{ steps.generate-token.outputs.token }}
+          release-type: node
+          config-file: release-please-config.json
+          manifest-file: .release-please-manifest.json
+    outputs:
+      release_created: ${{ steps.release.outputs.release_created }}
+      tag_name: ${{ steps.release.outputs.tag_name }}
+
  build-binaries:
    name: Build (${{ matrix.target }})
    runs-on: ubuntu-latest
+    needs: [release-please]
+    if: ${{ needs.release-please.outputs.release_created }}
    permissions:
      contents: read
    timeout-minutes: 15
@@ -43,9 +73,9 @@ jobs:
            archive: claude-agent-acp-windows-arm64.zip

    steps:
-      - uses: actions/checkout@08c6903cd8c0fde910a37f88322edcfb5dd907a8
+      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd

-      - uses: oven-sh/setup-bun@3d267786b128fe76c2f16a390aa2448b815359f3
+      - uses: oven-sh/setup-bun@0c5077e51419868618aeaa5fe8019c62421857d6
        with:
          bun-version: latest

@@ -69,7 +99,7 @@ jobs:
          fi

      - name: Upload artifact
-        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
+        uses: actions/upload-artifact@bbbca2ddaa5d8feaa63e36b76fdaad77386f024f
        with:
          name: ${{ matrix.archive }}
          path: ${{ matrix.archive }}
@@ -77,16 +107,16 @@ jobs:
  publish-npm:
    name: Publish to npm
    runs-on: ubuntu-latest
-    needs: [build-binaries]
-    if: startsWith(github.ref, 'refs/tags/')
+    needs: [build-binaries, release-please]
+    if: ${{ needs.release-please.outputs.release_created }}
    environment: release # Optional: for enhanced security
    permissions:
      contents: read
      id-token: write
    steps:
-      - uses: actions/checkout@08c6903cd8c0fde910a37f88322edcfb5dd907a8
+      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd
      # Setup .npmrc file to publish to npm
-      - uses: actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020
+      - uses: actions/setup-node@53b83947a5a98c8d113130e565377fae1a50d02f
        with:
          node-version: "lts/*"
          registry-url: "https://registry.npmjs.org"
@@ -99,22 +129,26 @@ jobs:
  release:
    name: Create Release
    runs-on: ubuntu-latest
-    needs: [build-binaries, publish-npm]
-    if: startsWith(github.ref, 'refs/tags/')
+    needs: [build-binaries, publish-npm, release-please]
+    if: ${{ needs.release-please.outputs.release_created }}
    permissions:
      contents: write

    steps:
-      - uses: actions/checkout@08c6903cd8c0fde910a37f88322edcfb5dd907a8
+      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd

      - name: Download all artifacts
-        uses: actions/download-artifact@37930b1c2abaa49bbe596cd826c3c89aef350131
+        uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c
        with:
          path: artifacts
          merge-multiple: true

-      - name: Create Release
-        uses: softprops/action-gh-release@6da8fa9354ddfdc4aeace5fc48d7f679b5214090
-        with:
-          generate_release_notes: true
-          files: artifacts/*
+      - name: Upload artifacts
+        shell: bash
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: |
+          gh release upload \
+            "${{ needs.release-please.outputs.tag_name }}" \
+            artifacts/* \
+            --clobber
--- a/.prettierignore
+++ b/.prettierignore
@@ -1,3 +1,7 @@
 node_modules
 dist
 .agent-shell
+
+.release-please-manifest.json
+release-please-config.json
+CHANGELOG.md
--- a/.release-please-manifest.json
+++ b/.release-please-manifest.json
@@ -0,0 +1,3 @@
+{
+  ".": "0.25.2"
+}
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,5 +1,55 @@
 # Changelog

+## [0.25.2](https://github.com/agentclientprotocol/claude-agent-acp/compare/v0.25.1...v0.25.2) (2026-04-06)
+
+
+### Bug Fixes
+
+* prioritize ANTHROPIC_MODEL env var over settings.model in model … ([#505](https://github.com/agentclientprotocol/claude-agent-acp/issues/505)) ([bea1a40](https://github.com/agentclientprotocol/claude-agent-acp/commit/bea1a40e9bfe1e06672b118a727f9339def3be23))
+
+## [0.25.1](https://github.com/agentclientprotocol/claude-agent-acp/compare/v0.25.0...v0.25.1) (2026-04-06)
+
+
+### Bug Fixes
+
+* add `auto` to valid modes in applySessionMode to fix mode cycling ([#507](https://github.com/agentclientprotocol/claude-agent-acp/issues/507)) ([15e91fb](https://github.com/agentclientprotocol/claude-agent-acp/commit/15e91fb5c3449de2600b583cb7a8d36b5b510443))
+
+## [0.25.0](https://github.com/agentclientprotocol/claude-agent-acp/compare/v0.24.2...v0.25.0) (2026-04-03)
+
+
+### Features
+
+* Add auto permission mode support ([#501](https://github.com/agentclientprotocol/claude-agent-acp/issues/501)) ([a161453](https://github.com/agentclientprotocol/claude-agent-acp/commit/a16145396fe2e6ead8734478961b2de65707e335))
+* Add separate Claude and Console terminal logins ([#502](https://github.com/agentclientprotocol/claude-agent-acp/issues/502)) ([063cd35](https://github.com/agentclientprotocol/claude-agent-acp/commit/063cd353809df8f3dda9e57d8ea848c0a6d44257))
+* Update to claude-agent-sdk 0.2.91 ([#500](https://github.com/agentclientprotocol/claude-agent-acp/issues/500)) ([65a2230](https://github.com/agentclientprotocol/claude-agent-acp/commit/65a223038576d72b74e1483fed10e982a1f842bd))
+
+
+### Bug Fixes
+
+* log warnings for malformed settings files instead of silent fallback ([#486](https://github.com/agentclientprotocol/claude-agent-acp/issues/486)) ([ae6c388](https://github.com/agentclientprotocol/claude-agent-acp/commit/ae6c38831415f9fc1de2d3dd1d4a247becbbd32f))
+* prevent race conditions in SettingsManager setCwd and debounce ([#485](https://github.com/agentclientprotocol/claude-agent-acp/issues/485)) ([7506223](https://github.com/agentclientprotocol/claude-agent-acp/commit/7506223cffb1aba4b4560feda11f69a1395a8c9d))
+* use current model's context window for usage_update size ([#412](https://github.com/agentclientprotocol/claude-agent-acp/issues/412)) ([d07799d](https://github.com/agentclientprotocol/claude-agent-acp/commit/d07799d7b3b4e438c8b158266c79723a0b592c07))
+
+## [0.24.2](https://github.com/agentclientprotocol/claude-agent-acp/compare/v0.24.1...v0.24.2) (2026-03-27)
+
+
+### Bug Fixes
+
+* Add explicit type checks for MCP servers (http/sse) ([#487](https://github.com/agentclientprotocol/claude-agent-acp/issues/487)) ([e00a439](https://github.com/agentclientprotocol/claude-agent-acp/commit/e00a43901fa2b4fd7d582e1de57277be88233007))
+
+## [0.24.1](https://github.com/agentclientprotocol/claude-agent-acp/compare/v0.24.0...v0.24.1) (2026-03-26)
+
+
+### Bug Fixes
+
+* Cleanup based on new idle state [#463](https://github.com/agentclientprotocol/claude-agent-acp/issues/463) ([#480](https://github.com/agentclientprotocol/claude-agent-acp/issues/480)) ([23b3073](https://github.com/agentclientprotocol/claude-agent-acp/commit/23b30730253752f0bc4e30b619a6236f16fafdb9))
+
+## 0.24.0
+
+Rename from `@zed-industries/claude-agent-acp` to `@agentclientprotocol/claude-agent-acp`.
+
+We are moving this to the main ACP org to better allow multiple teams to contribute and maintain this adapter.
+
 ## 0.23.1

 - Add back error_during_execution break point (#469)
--- a/61
+++ b/61
@@ -1,42 +1,18 @@
-Copyright 2022 - 2025 Zed Industries, Inc.

-
-   Licensed under the Apache License, Version 2.0 (the "License");
-   you may not use this file except in compliance with the License.
-   You may obtain a copy of the License at
-
-
-       http://www.apache.org/licenses/LICENSE-2.0
-
-
-   Unless required by applicable law or agreed to in writing, software
-   distributed under the License is distributed on an "AS IS" BASIS,
-   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-   See the License for the specific language governing permissions and
-   limitations under the License.
-
-
-
-
-Apache License
+                                 Apache License
                           Version 2.0, January 2004
                        http://www.apache.org/licenses/

-
   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION

-
   1. Definitions.

-
      "License" shall mean the terms and conditions for use, reproduction,
      and distribution as defined by Sections 1 through 9 of this document.

-
      "Licensor" shall mean the copyright owner or entity authorized by
      the copyright owner that is granting the License.

-
      "Legal Entity" shall mean the union of the acting entity and all
      other entities that control, are controlled by, or are under common
      control with that entity. For the purposes of this definition,
@@ -45,28 +21,23 @@ Apache License
      otherwise, or (ii) ownership of fifty percent (50%) or more of the
      outstanding shares, or (iii) beneficial ownership of such entity.

-
      "You" (or "Your") shall mean an individual or Legal Entity
      exercising permissions granted by this License.

-
      "Source" form shall mean the preferred form for making modifications,
      including but not limited to software source code, documentation
      source, and configuration files.

-
      "Object" form shall mean any form resulting from mechanical
      transformation or translation of a Source form, including but
      not limited to compiled object code, generated documentation,
      and conversions to other media types.

-
      "Work" shall mean the work of authorship, whether in Source or
      Object form, made available under the License, as indicated by a
      copyright notice that is included in or attached to the work
      (an example is provided in the Appendix below).

-
      "Derivative Works" shall mean any work, whether in Source or Object
      form, that is based on (or derived from) the Work and for which the
      editorial revisions, annotations, elaborations, or other modifications
@@ -75,7 +46,6 @@ Apache License
      separable from, or merely link (or bind by name) to the interfaces of,
      the Work and Derivative Works thereof.

-
      "Contribution" shall mean any work of authorship, including
      the original version of the Work and any modifications or additions
      to that Work or Derivative Works thereof, that is intentionally
@@ -90,12 +60,10 @@ Apache License
      excluding communication that is conspicuously marked or otherwise
      designated in writing by the copyright owner as "Not a Contribution."

-
      "Contributor" shall mean Licensor and any individual or Legal Entity
      on behalf of whom a Contribution has been received by Licensor and
      subsequently incorporated within the Work.

-
   2. Grant of Copyright License. Subject to the terms and conditions of
      this License, each Contributor hereby grants to You a perpetual,
      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
@@ -103,7 +71,6 @@ Apache License
      publicly display, publicly perform, sublicense, and distribute the
      Work and such Derivative Works in Source or Object form.

-
   3. Grant of Patent License. Subject to the terms and conditions of
      this License, each Contributor hereby grants to You a perpetual,
      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
@@ -120,28 +87,23 @@ Apache License
      granted to You under this License for that Work shall terminate
      as of the date such litigation is filed.

-
   4. Redistribution. You may reproduce and distribute copies of the
      Work or Derivative Works thereof in any medium, with or without
      modifications, and in Source or Object form, provided that You
      meet the following conditions:

-
      (a) You must give any other recipients of the Work or
          Derivative Works a copy of this License; and

-
      (b) You must cause any modified files to carry prominent notices
          stating that You changed the files; and

-
      (c) You must retain, in the Source form of any Derivative Works
          that You distribute, all copyright, patent, trademark, and
          attribution notices from the Source form of the Work,
          excluding those notices that do not pertain to any part of
          the Derivative Works; and

-
      (d) If the Work includes a "NOTICE" text file as part of its
          distribution, then any Derivative Works that You distribute must
          include a readable copy of the attribution notices contained
@@ -159,7 +121,6 @@ Apache License
          that such additional attribution notices cannot be construed
          as modifying the License.

-
      You may add Your own copyright statement to Your modifications and
      may provide additional or different license terms and conditions
      for use, reproduction, or distribution of Your modifications, or
@@ -167,7 +128,6 @@ Apache License
      reproduction, and distribution of the Work otherwise complies with
      the conditions stated in this License.

-
   5. Submission of Contributions. Unless You explicitly state otherwise,
      any Contribution intentionally submitted for inclusion in the Work
      by You to the Licensor shall be under the terms and conditions of
@@ -176,13 +136,11 @@ Apache License
      the terms of any separate license agreement you may have executed
      with Licensor regarding such Contributions.

-
   6. Trademarks. This License does not grant permission to use the trade
      names, trademarks, service marks, or product names of the Licensor,
      except as required for reasonable and customary use in describing the
      origin of the Work and reproducing the content of the NOTICE file.

-
   7. Disclaimer of Warranty. Unless required by applicable law or
      agreed to in writing, Licensor provides the Work (and each
      Contributor provides its Contributions) on an "AS IS" BASIS,
@@ -193,7 +151,6 @@ Apache License
      appropriateness of using or redistributing the Work and assume any
      risks associated with Your exercise of permissions under this License.

-
   8. Limitation of Liability. In no event and under no legal theory,
      whether in tort (including negligence), contract, or otherwise,
      unless required by applicable law (such as deliberate and grossly
@@ -206,7 +163,6 @@ Apache License
      other commercial damages or losses), even if such Contributor
      has been advised of the possibility of such damages.

-
   9. Accepting Warranty or Additional Liability. While redistributing
      the Work or Derivative Works thereof, You may choose to offer,
      and charge a fee for, acceptance of support, warranty, indemnity,
@@ -218,5 +174,18 @@ Apache License
      incurred by, or claims asserted against, such Contributor by reason
      of your accepting any such warranty or additional liability.

-
   END OF TERMS AND CONDITIONS
+
+   Copyright 2025 Zed Industries, Inc. and contributors
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.
--- a/README.md
+++ b/README.md
@@ -1,8 +1,8 @@
 # ACP adapter for the Claude Agent SDK

-[![npm](https://img.shields.io/npm/v/%40zed-industries%2Fclaude-agent-acp)](https://www.npmjs.com/package/@zed-industries/claude-agent-acp)
+[![npm](https://img.shields.io/npm/v/%40agentclientprotocol%2Fclaude-agent-acp)](https://www.npmjs.com/package/@agentclientprotocol/claude-agent-acp)

-Use [Claude Agent SDK](https://platform.claude.com/docs/en/agent-sdk/overview#branding-guidelines) from [ACP-compatible](https://agentclientprotocol.com) clients such as [Zed](https://zed.dev)!
+Use [Claude Agent SDK](https://platform.claude.com/docs/en/agent-sdk/overview#branding-guidelines) from [ACP-compatible](https://agentclientprotocol.com) clients!

 This tool implements an ACP agent by using the official [Claude Agent SDK](https://platform.claude.com/docs/en/agent-sdk/overview), supporting:

@@ -18,41 +18,8 @@ This tool implements an ACP agent by using the official [Claude Agent SDK](https

 Learn more about the [Agent Client Protocol](https://agentclientprotocol.com/).

-## How to use
+## Contribution Policy

-### Zed
+This project does not require a Contributor License Agreement (CLA). Instead, contributions are accepted under the following terms:

-The latest version of Zed can already use this adapter out of the box.
-
-To use Claude Agent, open the Agent Panel and click "New Claude Agent Thread" from the `+` button menu in the top-right:
-
-https://github.com/user-attachments/assets/ddce66c7-79ac-47a3-ad59-4a6a3ca74903
-
-Read the docs on [External Agent](https://zed.dev/docs/ai/external-agents) support.
-
-### Other clients
-
-Or try it with any of the other [ACP compatible clients](https://agentclientprotocol.com/overview/clients)!
-
-#### Installation
-
-Install the adapter from `npm`:
-
-```bash
-npm install -g @zed-industries/claude-agent-acp
-```
-
-You can then use `claude-agent-acp` as a regular ACP agent:
-
-```
-ANTHROPIC_API_KEY=sk-... claude-agent-acp
-```
-
-#### Single-file executable
-
-Pre-built single-file binaries are available on the [Releases](https://github.com/zed-industries/claude-agent-acp/releases) page for Linux, macOS, and Windows.
-These binaries bundle everything needed and don't require Node.js.
-
-## License
-
-Apache-2.0
+> By contributing to this project, you agree that your contributions will be licensed under the [Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-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).
--- a/package-lock.json
+++ b/package-lock.json
--- a/package.json
+++ b/package.json
@@ -1,9 +1,9 @@
 {
-  "name": "@zed-industries/claude-agent-acp",
+  "name": "@agentclientprotocol/claude-agent-acp",
  "publishConfig": {
    "access": "public"
  },
-  "version": "0.23.1",
+  "version": "0.25.2",
  "description": "An ACP-compatible coding agent powered by the Claude Agent SDK (TypeScript)",
  "main": "dist/lib.js",
  "types": "dist/lib.d.ts",
@@ -49,33 +49,33 @@
    "sdk",
    "code"
  ],
-  "homepage": "https://github.com/zed-industries/claude-agent-acp#readme",
+  "homepage": "https://github.com/agentclientprotocol/claude-agent-acp#readme",
  "bugs": {
-    "url": "https://github.com/zed-industries/claude-agent-acp/issues"
+    "url": "https://github.com/agentclientprotocol/claude-agent-acp/issues"
  },
  "repository": {
    "type": "git",
-    "url": "git+https://github.com/zed-industries/claude-agent-acp.git"
+    "url": "git+https://github.com/agentclientprotocol/claude-agent-acp.git"
  },
  "author": "Zed Industries",
  "license": "Apache-2.0",
  "dependencies": {
-    "@agentclientprotocol/sdk": "0.17.0",
-    "@anthropic-ai/claude-agent-sdk": "0.2.83",
+    "@agentclientprotocol/sdk": "0.18.0",
+    "@anthropic-ai/claude-agent-sdk": "0.2.92",
    "zod": "^3.25.0 || ^4.0.0"
  },
  "devDependencies": {
-    "@anthropic-ai/sdk": "0.80.0",
+    "@anthropic-ai/sdk": "0.82.0",
    "@eslint/js": "10.0.1",
-    "@types/node": "25.5.0",
-    "@typescript-eslint/eslint-plugin": "8.57.2",
-    "@typescript-eslint/parser": "8.57.2",
-    "eslint": "10.1.0",
+    "@types/node": "25.5.2",
+    "@typescript-eslint/eslint-plugin": "8.58.0",
+    "@typescript-eslint/parser": "8.58.0",
+    "eslint": "10.2.0",
    "eslint-config-prettier": "10.1.8",
    "globals": "17.4.0",
    "prettier": "3.8.1",
    "ts-node": "10.9.2",
-    "typescript": "5.9.3",
-    "vitest": "4.1.1"
+    "typescript": "6.0.2",
+    "vitest": "4.1.2"
  }
 }
--- a/release-please-config.json
+++ b/release-please-config.json
@@ -0,0 +1,13 @@
+{
+  "bump-minor-pre-major": true,
+  "bump-patch-for-minor-pre-major": true,
+  "packages": {
+    ".": {
+      "changelog-path": "CHANGELOG.md",
+      "release-type": "node",
+      "bump-minor-pre-major": true,
+      "bump-patch-for-minor-pre-major": true
+    }
+  },
+  "$schema": "https://raw.githubusercontent.com/googleapis/release-please/main/schemas/config.json"
+}
--- a/src/acp-agent.ts
+++ b/src/acp-agent.ts
@@ -40,6 +40,7 @@ import {
  TerminalOutputResponse,
  WriteTextFileRequest,
  WriteTextFileResponse,
+  StopReason,
 } from "@agentclientprotocol/sdk";
 import {
  CanUseTool,
@@ -47,6 +48,7 @@ import {
  listSessions,
  McpServerConfig,
  ModelInfo,
+  ModelUsage,
  Options,
  PermissionMode,
  Query,
@@ -232,6 +234,7 @@ const ALLOW_BYPASS = !IS_ROOT || !!process.env.IS_SANDBOX;
 const LOCAL_ONLY_COMMANDS = new Set(["/context", "/heapdump", "/extra-usage"]);

 const PERMISSION_MODE_ALIASES: Record<string, PermissionMode> = {
+  auto: "auto",
  default: "default",
  acceptedits: "acceptEdits",
  dontask: "dontAsk",
@@ -305,25 +308,42 @@ export class ClaudeAcpAgent implements Agent {
      },
    };

-    const terminalAuthMethod: any = {
-      description: "Run `claude /login` in the terminal",
-      name: "Log in with Claude",
-      id: "claude-login",
-      type: "terminal",
-      args: ["--cli"],
-    };
    const supportsTerminalAuth = request.clientCapabilities?.auth?.terminal === true;
+    const supportsMetaTerminalAuth = request.clientCapabilities?._meta?.["terminal-auth"] === true;
+
+    const claudeLoginMethod: any = {
+      description: "Use Claude subscription ",
+      name: "Claude Subscription",
+      id: "claude-ai-login",
+      type: "terminal",
+      args: ["--cli", "auth", "login", "--claudeai"],
+    };
+
+    const consoleLoginMethod: any = {
+      description: "Use Anthropic Console (API usage billing)",
+      name: "Anthropic Console",
+      id: "console-login",
+      type: "terminal",
+      args: ["--cli", "auth", "login", "--console"],
+    };

    // If client supports terminal-auth capability, use that instead.
-    const supportsMetaTerminalAuth = request.clientCapabilities?._meta?.["terminal-auth"] === true;
    if (supportsMetaTerminalAuth) {
-      terminalAuthMethod._meta = {
+      const baseArgs = process.argv.slice(1);
+      claudeLoginMethod._meta = {
        "terminal-auth": {
          command: process.execPath,
-          args: [...process.argv.slice(1), "--cli"],
+          args: [...baseArgs, "--cli", "auth", "login", "--claudeai"],
          label: "Claude Login",
        },
      };
+      consoleLoginMethod._meta = {
+        "terminal-auth": {
+          command: process.execPath,
+          args: [...baseArgs, "--cli", "auth", "login", "--console"],
+          label: "Anthropic Console Login",
+        },
+      };
    }

    return {
@@ -357,8 +377,9 @@ export class ClaudeAcpAgent implements Agent {
      },
      authMethods: [
        ...(!shouldHideClaudeAuth() && (supportsTerminalAuth || supportsMetaTerminalAuth)
-          ? [terminalAuthMethod]
+          ? [claudeLoginMethod]
          : []),
+        ...(supportsTerminalAuth || supportsMetaTerminalAuth ? [consoleLoginMethod] : []),
        ...(supportsGatewayAuth ? [gatewayAuthMethod] : []),
      ],
    };
@@ -467,23 +488,20 @@ export class ClaudeAcpAgent implements Agent {
    };

    let lastAssistantTotalUsage: number | null = null;
+    let lastAssistantModel: string | null = null;
+    let lastContextWindowSize: number = 200000;

    const userMessage = promptToClaude(params);

    const promptUuid = randomUUID();
    userMessage.uuid = promptUuid;

-    let promptReplayed = false;
-
    // These local-only commands return a result without replaying the user
    // message. Mark promptReplayed=true so their result isn't consumed as a
    // background task result.
    const firstText = params.prompt[0]?.type === "text" ? params.prompt[0].text : "";
    const isLocalOnlyCommand =
      firstText.startsWith("/") && LOCAL_ONLY_COMMANDS.has(firstText.split(" ", 1)[0]);
-    if (isLocalOnlyCommand) {
-      promptReplayed = true;
-    }

    if (session.promptRunning) {
      session.input.push(userMessage);
@@ -494,15 +512,13 @@ export class ClaudeAcpAgent implements Agent {
      if (cancelled) {
        return { stopReason: "cancelled" };
      }
-      // The replay resolved the promise, mark in this loop too,
-      // so we don't treat the next result as a background task's result.
-      promptReplayed = true;
    } else {
      session.input.push(userMessage);
    }

    session.promptRunning = true;
    let handedOff = false;
+    let stopReason: StopReason = "end_turn";

    try {
      while (true) {
@@ -533,9 +549,26 @@ export class ClaudeAcpAgent implements Agent {
                break;
              }
              case "compact_boundary": {
-                // We don't know the exact size, but since we compacted,
-                // we set it to zero. The client gets the exact size on the next message.
+                // Send used:0 immediately so the client doesn't keep showing
+                // the stale pre-compaction context size until the next turn.
+                //
+                // This is a deliberate approximation: we don't know the exact
+                // post-compaction token count (only the SDK's next API call
+                // reveals that). But used:0 is directionally correct — context
+                // just dropped dramatically — and the real value replaces it
+                // within seconds when the next result message arrives.
+                // The alternative (no update) leaves the client showing e.g.
+                // "944k/1m" right after the user sees "Compacting completed",
+                // which is confusing and wrong.
                lastAssistantTotalUsage = 0;
+                await this.client.sessionUpdate({
+                  sessionId: message.session_id,
+                  update: {
+                    sessionUpdate: "usage_update",
+                    used: 0,
+                    size: lastContextWindowSize,
+                  },
+                });
                await this.client.sessionUpdate({
                  sessionId: message.session_id,
                  update: {
@@ -543,7 +576,6 @@ export class ClaudeAcpAgent implements Agent {
                    content: { type: "text", text: "\n\nCompacting completed." },
                  },
                });
-                promptReplayed = true;
                break;
              }
              case "local_command_output": {
@@ -554,12 +586,11 @@ export class ClaudeAcpAgent implements Agent {
                    content: { type: "text", text: message.content },
                  },
                });
-                promptReplayed = true;
                break;
              }
              case "session_state_changed": {
                if (message.state === "idle") {
-                  return { stopReason: "end_turn", usage: sessionUsage(session) };
+                  return { stopReason, usage: sessionUsage(session) };
                }
                break;
              }
@@ -586,10 +617,11 @@ export class ClaudeAcpAgent implements Agent {
            session.accumulatedUsage.cachedReadTokens += message.usage.cache_read_input_tokens;
            session.accumulatedUsage.cachedWriteTokens += message.usage.cache_creation_input_tokens;

-            // Calculate context window size from modelUsage (minimum across all models used)
-            const contextWindows = Object.values(message.modelUsage).map((m) => m.contextWindow);
-            const contextWindowSize =
-              contextWindows.length > 0 ? Math.min(...contextWindows) : 200000;
+            const matchingModelUsage = lastAssistantModel
+              ? getMatchingModelUsage(message.modelUsage, lastAssistantModel)
+              : null;
+            const contextWindowSize = matchingModelUsage?.contextWindow ?? 200000;
+            lastContextWindowSize = contextWindowSize;

            // Send usage_update notification
            if (lastAssistantTotalUsage !== null) {
@@ -607,32 +639,19 @@ export class ClaudeAcpAgent implements Agent {
              });
            }

-            // Check cancelled before promptReplayed — when a cancel races
-            // with the first result, promptReplayed is still false and the
-            // result would be consumed as a background task, blocking the
-            // loop forever (see #442).
            if (session.cancelled) {
-              return { stopReason: "cancelled" };
-            }
-
-            if (!promptReplayed) {
-              // This result is from a background task that finished after
-              // the previous prompt loop ended. Consume it and continue
-              // waiting for our own prompt's result.
-              this.logger.log(`Session ${params.sessionId}: consuming background task result`);
+              stopReason = "cancelled";
              break;
            }

-            // Build the usage response
-            const usage = sessionUsage(session);
-
            switch (message.subtype) {
              case "success": {
                if (message.result.includes("Please run /login")) {
                  throw RequestError.authRequired();
                }
                if (message.stop_reason === "max_tokens") {
-                  return { stopReason: "max_tokens", usage };
+                  stopReason = "max_tokens";
+                  break;
                }
                if (message.is_error) {
                  throw RequestError.internalError(undefined, message.result);
@@ -655,7 +674,8 @@ export class ClaudeAcpAgent implements Agent {
              }
              case "error_during_execution": {
                if (message.stop_reason === "max_tokens") {
-                  return { stopReason: "max_tokens", usage };
+                  stopReason = "max_tokens";
+                  break;
                }
                if (message.is_error) {
                  throw RequestError.internalError(
@@ -663,7 +683,8 @@ export class ClaudeAcpAgent implements Agent {
                    message.errors.join(", ") || message.subtype,
                  );
                }
-                return { stopReason: "end_turn", usage: sessionUsage(session) };
+                stopReason = "end_turn";
+                break;
              }
              case "error_max_budget_usd":
              case "error_max_turns":
@@ -674,7 +695,8 @@ export class ClaudeAcpAgent implements Agent {
                    message.errors.join(", ") || message.subtype,
                  );
                }
-                return { stopReason: "max_turn_requests", usage };
+                stopReason = "max_turn_requests";
+                break;
              default:
                unreachable(message, this.logger);
                break;
@@ -706,9 +728,6 @@ export class ClaudeAcpAgent implements Agent {
            // Check for prompt replay
            if (message.type === "user" && "uuid" in message && message.uuid) {
              if (message.uuid === promptUuid) {
-                // Our own prompt was replayed back — we're now processing
-                // our prompt's response (not a background task's).
-                promptReplayed = true;
                break;
              }

@@ -728,6 +747,11 @@ export class ClaudeAcpAgent implements Agent {
            }

            // Store latest assistant usage (excluding subagents)
+            // Sum all token types as a proxy for post-turn context occupancy:
+            // current turn's output will become next turn's input.
+            // Note: per the Anthropic API, input_tokens excludes cache tokens —
+            // cache_read and cache_creation are reported separately, so summing
+            // all four fields is not double-counting.
            if ((message.message as any).usage && message.parent_tool_use_id === null) {
              const messageWithUsage = message.message as unknown as SDKResultMessage;
              lastAssistantTotalUsage =
@@ -736,6 +760,16 @@ export class ClaudeAcpAgent implements Agent {
                messageWithUsage.usage.cache_read_input_tokens +
                messageWithUsage.usage.cache_creation_input_tokens;
            }
+            // Track the current top-level model for context window size lookup
+            // (exclude subagent messages to stay in sync with lastAssistantTotalUsage)
+            if (
+              message.type === "assistant" &&
+              message.parent_tool_use_id === null &&
+              message.message.model &&
+              message.message.model !== "<synthetic>"
+            ) {
+              lastAssistantModel = message.message.model;
+            }

            // Slash commands like /compact can generate invalid output... doesn't match
            // their own docs: https://docs.anthropic.com/en/docs/claude-code/sdk/sdk-slash-commands#%2Fcompact-compact-conversation-history
@@ -972,6 +1006,7 @@ export class ClaudeAcpAgent implements Agent {

  private async applySessionMode(sessionId: string, modeId: string): Promise<void> {
    switch (modeId) {
+      case "auto":
      case "default":
      case "acceptEdits":
      case "bypassPermissions":
@@ -1277,7 +1312,8 @@ export class ClaudeAcpAgent implements Agent {
    const mcpServers: Record<string, McpServerConfig> = {};
    if (Array.isArray(params.mcpServers)) {
      for (const server of params.mcpServers) {
-        if ("type" in server) {
+        if ("type" in server && (server.type === "http" || server.type === "sse")) {
+          // HTTP or SSE type MCP server
          mcpServers[server.name] = {
            type: server.type,
            url: server.url,
@@ -1286,6 +1322,7 @@ export class ClaudeAcpAgent implements Agent {
              : undefined,
          };
        } else {
+          // Stdio type MCP server (with or without explicit type field)
          mcpServers[server.name] = {
            type: "stdio",
            command: server.command,
@@ -1447,6 +1484,11 @@ export class ClaudeAcpAgent implements Agent {
    const models = await getAvailableModels(q, initializationResult.models, settingsManager);

    const availableModes = [
+      {
+        id: "auto",
+        name: "Auto",
+        decription: "Use a model classifier to approve/deny permission prompts.",
+      },
      {
        id: "default",
        name: "Default",
@@ -1657,7 +1699,16 @@ async function getAvailableModels(

  let currentModel = models[0];

-  if (settings.model) {
+  // Model priority (highest to lowest):
+  // 1. ANTHROPIC_MODEL environment variable
+  // 2. settings.model (user configuration)
+  // 3. models[0] (default first model)
+  if (process.env.ANTHROPIC_MODEL) {
+    const match = resolveModelPreference(models, process.env.ANTHROPIC_MODEL);
+    if (match) {
+      currentModel = match;
+    }
+  } else if (settings.model) {
    const match = resolveModelPreference(models, settings.model);
    if (match) {
      currentModel = match;
@@ -1735,10 +1786,10 @@ export function promptToClaude(prompt: PromptRequest): SDKUserMessage {
      case "text": {
        let text = chunk.text;
        // change /mcp:server:command args -> /server:command (MCP) args
-        const mcpMatch = text.match(/^\/mcp:([^:\s]+):(\S+)(\s+.*)?$/);
+        const mcpMatch = text.match(/^\/mcp:([^:\s]+):(\S+)(?:\s(.*))?$/);
        if (mcpMatch) {
          const [, server, command, args] = mcpMatch;
-          text = `/${server}:${command} (MCP)${args || ""}`;
+          text = `/${server}:${command} (MCP)${args ? ` ${args}` : ""}`;
        }
        content.push({ type: "text", text });
        break;
@@ -2128,3 +2179,28 @@ export function runAcp() {
  const stream = ndJsonStream(input, output);
  new AgentSideConnection((client) => new ClaudeAcpAgent(client), stream);
 }
+
+function commonPrefixLength(a: string, b: string) {
+  let i = 0;
+  while (i < a.length && i < b.length && a[i] === b[i]) {
+    i++;
+  }
+  return i;
+}
+
+function getMatchingModelUsage(modelUsage: Record<string, ModelUsage>, currentModel: string) {
+  let bestKey: string | null = null;
+  let bestLen = 0;
+
+  for (const key of Object.keys(modelUsage)) {
+    const len = commonPrefixLength(key, currentModel);
+    if (len > bestLen) {
+      bestLen = len;
+      bestKey = key;
+    }
+  }
+
+  if (bestKey) {
+    return modelUsage[bestKey];
+  }
+}
--- a/src/settings.ts
+++ b/src/settings.ts
@@ -25,9 +25,14 @@ export interface ClaudeCodeSettings {
 }

 /**
- * Reads and parses a JSON settings file, returning an empty object if not found or invalid
+ * Reads and parses a JSON settings file, returning an empty object if not found or invalid.
+ * Silently ignores missing files (ENOENT) but logs warnings for other errors
+ * (malformed JSON, permission errors, etc.) to aid debugging.
 */
-async function loadSettingsFile(filePath: string | null): Promise<ClaudeCodeSettings> {
+async function loadSettingsFile(
+  filePath: string | null,
+  logger?: { error: (...args: any[]) => void },
+): Promise<ClaudeCodeSettings> {
  if (!filePath) {
    return {};
  }
@@ -35,7 +40,11 @@ async function loadSettingsFile(filePath: string | null): Promise<ClaudeCodeSett
  try {
    const content = await fs.promises.readFile(filePath, "utf-8");
    return JSON.parse(content) as ClaudeCodeSettings;
-  } catch {
+  } catch (error) {
+    if (error instanceof Error && "code" in error && error.code === "ENOENT") {
+      return {};
+    }
+    logger?.error(`Failed to load settings from ${filePath}:`, error);
    return {};
  }
 }
@@ -83,7 +92,9 @@ export class SettingsManager {
  private onChange?: () => void;
  private logger: { log: (...args: any[]) => void; error: (...args: any[]) => void };
  private initialized = false;
+  private disposed = false;
  private debounceTimer: ReturnType<typeof setTimeout> | null = null;
+  private initPromise: Promise<void> | null = null;

  constructor(cwd: string, options?: SettingsManagerOptions) {
    this.cwd = cwd;
@@ -98,10 +109,19 @@ export class SettingsManager {
    if (this.initialized) {
      return;
    }
+    if (this.initPromise) {
+      return this.initPromise;
+    }

-    await this.loadAllSettings();
-    this.setupWatchers();
-    this.initialized = true;
+    this.disposed = false;
+    this.initPromise = this.loadAllSettings().then(() => {
+      if (!this.disposed) {
+        this.setupWatchers();
+        this.initialized = true;
+      }
+      this.initPromise = null;
+    });
+    return this.initPromise;
  }

  /**
@@ -130,10 +150,10 @@ export class SettingsManager {
   */
  private async loadAllSettings(): Promise<void> {
    const [userSettings, projectSettings, localSettings, enterpriseSettings] = await Promise.all([
-      loadSettingsFile(this.getUserSettingsPath()),
-      loadSettingsFile(this.getProjectSettingsPath()),
-      loadSettingsFile(this.getLocalSettingsPath()),
-      loadSettingsFile(getManagedSettingsPath()),
+      loadSettingsFile(this.getUserSettingsPath(), this.logger),
+      loadSettingsFile(this.getProjectSettingsPath(), this.logger),
+      loadSettingsFile(this.getLocalSettingsPath(), this.logger),
+      loadSettingsFile(getManagedSettingsPath(), this.logger),
    ]);

    this.userSettings = userSettings;
@@ -226,9 +246,14 @@ export class SettingsManager {

    this.debounceTimer = setTimeout(async () => {
      this.debounceTimer = null;
+      if (this.disposed) {
+        return;
+      }
      try {
        await this.loadAllSettings();
-        this.onChange?.();
+        if (!this.disposed) {
+          this.onChange?.();
+        }
      } catch (error) {
        this.logger.error("Failed to reload settings:", error);
      }
@@ -259,7 +284,6 @@ export class SettingsManager {

    this.dispose();
    this.cwd = cwd;
-    this.initialized = false;
    await this.initialize();
  }

@@ -267,6 +291,10 @@ export class SettingsManager {
   * Disposes of file watchers and cleans up resources
   */
  dispose(): void {
+    this.disposed = true;
+    this.initialized = false;
+    this.initPromise = null;
+
    if (this.debounceTimer) {
      clearTimeout(this.debounceTimer);
      this.debounceTimer = null;
@@ -276,6 +304,5 @@ export class SettingsManager {
      watcher.close();
    }
    this.watchers = [];
-    this.initialized = false;
  }
 }
--- a/src/tests/acp-agent.test.ts
+++ b/src/tests/acp-agent.test.ts
@@ -1355,6 +1355,7 @@ describe("stop reason propagation", () => {
    const agent = createMockAgent();
    injectSession(agent, [
      createResultMessage({ subtype: "success", stop_reason: "max_tokens", is_error: false }),
+      { type: "system", subtype: "session_state_changed", state: "idle" },
    ]);

    const response = await agent.prompt({
@@ -1374,6 +1375,7 @@ describe("stop reason propagation", () => {
        is_error: true,
        result: "Token limit reached",
      }),
+      { type: "system", subtype: "session_state_changed", state: "idle" },
    ]);

    const response = await agent.prompt({
@@ -1393,6 +1395,7 @@ describe("stop reason propagation", () => {
        is_error: true,
        errors: ["some error"],
      }),
+      { type: "system", subtype: "session_state_changed", state: "idle" },
    ]);

    const response = await agent.prompt({
@@ -1605,3 +1608,446 @@ describe("session/close", () => {
    expect(agent.sessions["session-b"]).toBeDefined();
  });
 });
+
+describe("usage_update computation", () => {
+  function createAssistantMessage(overrides: {
+    model: string;
+    usage?: {
+      input_tokens: number;
+      output_tokens: number;
+      cache_read_input_tokens: number;
+      cache_creation_input_tokens: number;
+    };
+  }) {
+    return {
+      type: "assistant" as const,
+      parent_tool_use_id: null,
+      uuid: randomUUID(),
+      session_id: "test-session",
+      message: {
+        model: overrides.model,
+        content: [{ type: "text", text: "hello" }],
+        usage: overrides.usage ?? {
+          input_tokens: 100,
+          output_tokens: 50,
+          cache_read_input_tokens: 20,
+          cache_creation_input_tokens: 10,
+        },
+      },
+    };
+  }
+
+  function createResultMessageWithModel(overrides: {
+    modelUsage: Record<
+      string,
+      {
+        inputTokens: number;
+        outputTokens: number;
+        cacheReadInputTokens: number;
+        cacheCreationInputTokens: number;
+        webSearchRequests: number;
+        costUSD: number;
+        contextWindow: number;
+        maxOutputTokens: number;
+      }
+    >;
+  }) {
+    return {
+      type: "result" as const,
+      subtype: "success" as const,
+      stop_reason: "end_turn",
+      is_error: false,
+      result: "",
+      errors: [],
+      duration_ms: 0,
+      duration_api_ms: 0,
+      num_turns: 1,
+      total_cost_usd: 0.01,
+      usage: {
+        input_tokens: 10,
+        output_tokens: 5,
+        cache_read_input_tokens: 0,
+        cache_creation_input_tokens: 0,
+      },
+      modelUsage: overrides.modelUsage,
+      permission_denials: [],
+      uuid: randomUUID(),
+      session_id: "test-session",
+    };
+  }
+
+  function createMockAgentWithCapture() {
+    const updates: any[] = [];
+    const mockClient = {
+      sessionUpdate: async (notification: any) => {
+        updates.push(notification);
+      },
+    } as unknown as AgentSideConnection;
+    const agent = new ClaudeAcpAgent(mockClient, { log: () => {}, error: () => {} });
+    return { agent, updates };
+  }
+
+  function injectSession(agent: ClaudeAcpAgent, messages: any[]) {
+    const input = new Pushable<any>();
+    async function* messageGenerator() {
+      // Wait for the prompt to push its user message so we can replay it
+      const iter = input[Symbol.asyncIterator]();
+      const { value: userMessage, done } = await iter.next();
+      if (!done && userMessage) {
+        yield {
+          type: "user",
+          message: userMessage.message,
+          parent_tool_use_id: null,
+          uuid: userMessage.uuid,
+          session_id: "test-session",
+          isReplay: true,
+        };
+      }
+      yield* messages;
+    }
+    agent.sessions["test-session"] = {
+      query: messageGenerator() as any,
+      input,
+      cancelled: false,
+      cwd: "/test",
+      modes: {
+        currentModeId: "default",
+        availableModes: [],
+      },
+      models: {
+        currentModelId: "default",
+        availableModels: [],
+      },
+      settingsManager: {} as any,
+      accumulatedUsage: {
+        inputTokens: 0,
+        outputTokens: 0,
+        cachedReadTokens: 0,
+        cachedWriteTokens: 0,
+      },
+      configOptions: [],
+      promptRunning: false,
+      pendingMessages: new Map(),
+      nextPendingOrder: 0,
+      abortController: new AbortController(),
+    };
+  }
+
+  it("used sums all token types as post-turn context occupancy proxy", async () => {
+    const { agent, updates } = createMockAgentWithCapture();
+    injectSession(agent, [
+      createAssistantMessage({
+        model: "claude-opus-4-20250514",
+        usage: {
+          input_tokens: 1000,
+          output_tokens: 500,
+          cache_read_input_tokens: 200,
+          cache_creation_input_tokens: 100,
+        },
+      }),
+      createResultMessageWithModel({
+        modelUsage: {
+          "claude-opus-4-20250514": {
+            inputTokens: 1000,
+            outputTokens: 500,
+            cacheReadInputTokens: 200,
+            cacheCreationInputTokens: 100,
+            webSearchRequests: 0,
+            costUSD: 0.01,
+            contextWindow: 1000000,
+            maxOutputTokens: 16384,
+          },
+        },
+      }),
+      { type: "system", subtype: "session_state_changed", state: "idle" },
+    ]);
+
+    await agent.prompt({ sessionId: "test-session", prompt: [{ type: "text", text: "test" }] });
+
+    const usageUpdate = updates.find((u: any) => u.update?.sessionUpdate === "usage_update");
+    expect(usageUpdate).toBeDefined();
+    // used = input(1000) + output(500) + cache_read(200) + cache_creation(100) = 1800
+    expect(usageUpdate.update.used).toBe(1800);
+  });
+
+  it("size reflects the current model's context window, not min across all", async () => {
+    const { agent, updates } = createMockAgentWithCapture();
+    injectSession(agent, [
+      createAssistantMessage({ model: "claude-opus-4-20250514" }),
+      createResultMessageWithModel({
+        modelUsage: {
+          "claude-opus-4-20250514": {
+            inputTokens: 100,
+            outputTokens: 50,
+            cacheReadInputTokens: 20,
+            cacheCreationInputTokens: 10,
+            webSearchRequests: 0,
+            costUSD: 0.01,
+            contextWindow: 1000000,
+            maxOutputTokens: 16384,
+          },
+          "claude-sonnet-4-20250514": {
+            inputTokens: 50,
+            outputTokens: 25,
+            cacheReadInputTokens: 10,
+            cacheCreationInputTokens: 5,
+            webSearchRequests: 0,
+            costUSD: 0.005,
+            contextWindow: 200000,
+            maxOutputTokens: 16384,
+          },
+        },
+      }),
+      { type: "system", subtype: "session_state_changed", state: "idle" },
+    ]);
+
+    await agent.prompt({ sessionId: "test-session", prompt: [{ type: "text", text: "test" }] });
+
+    const usageUpdate = updates.find((u: any) => u.update?.sessionUpdate === "usage_update");
+    expect(usageUpdate).toBeDefined();
+    // size should be 1000000 (Opus), not 200000 (min of both)
+    expect(usageUpdate.update.size).toBe(1000000);
+  });
+
+  it("after model switch, size updates to the new model's window", async () => {
+    const { agent, updates } = createMockAgentWithCapture();
+    // Simulate: assistant on Sonnet with both models in modelUsage
+    injectSession(agent, [
+      createAssistantMessage({ model: "claude-sonnet-4-20250514" }),
+      createResultMessageWithModel({
+        modelUsage: {
+          "claude-opus-4-20250514": {
+            inputTokens: 100,
+            outputTokens: 50,
+            cacheReadInputTokens: 20,
+            cacheCreationInputTokens: 10,
+            webSearchRequests: 0,
+            costUSD: 0.01,
+            contextWindow: 1000000,
+            maxOutputTokens: 16384,
+          },
+          "claude-sonnet-4-20250514": {
+            inputTokens: 50,
+            outputTokens: 25,
+            cacheReadInputTokens: 10,
+            cacheCreationInputTokens: 5,
+            webSearchRequests: 0,
+            costUSD: 0.005,
+            contextWindow: 200000,
+            maxOutputTokens: 16384,
+          },
+        },
+      }),
+      { type: "system", subtype: "session_state_changed", state: "idle" },
+    ]);
+
+    await agent.prompt({ sessionId: "test-session", prompt: [{ type: "text", text: "test" }] });
+
+    const usageUpdate = updates.find((u: any) => u.update?.sessionUpdate === "usage_update");
+    expect(usageUpdate).toBeDefined();
+    // size should be 200000 (Sonnet - the current model)
+    expect(usageUpdate.update.size).toBe(200000);
+  });
+
+  it("after switching back to original model, size returns to original window", async () => {
+    const { agent, updates } = createMockAgentWithCapture();
+    // Last assistant message is Opus again
+    injectSession(agent, [
+      createAssistantMessage({ model: "claude-sonnet-4-20250514" }),
+      createAssistantMessage({ model: "claude-opus-4-20250514" }),
+      createResultMessageWithModel({
+        modelUsage: {
+          "claude-opus-4-20250514": {
+            inputTokens: 200,
+            outputTokens: 100,
+            cacheReadInputTokens: 40,
+            cacheCreationInputTokens: 20,
+            webSearchRequests: 0,
+            costUSD: 0.02,
+            contextWindow: 1000000,
+            maxOutputTokens: 16384,
+          },
+          "claude-sonnet-4-20250514": {
+            inputTokens: 50,
+            outputTokens: 25,
+            cacheReadInputTokens: 10,
+            cacheCreationInputTokens: 5,
+            webSearchRequests: 0,
+            costUSD: 0.005,
+            contextWindow: 200000,
+            maxOutputTokens: 16384,
+          },
+        },
+      }),
+      { type: "system", subtype: "session_state_changed", state: "idle" },
+    ]);
+
+    await agent.prompt({ sessionId: "test-session", prompt: [{ type: "text", text: "test" }] });
+
+    const usageUpdate = updates.find((u: any) => u.update?.sessionUpdate === "usage_update");
+    expect(usageUpdate).toBeDefined();
+    // size should be 1000000 (Opus - switched back)
+    expect(usageUpdate.update.size).toBe(1000000);
+  });
+
+  it("subagent assistant messages do not affect size (top-level model is used)", async () => {
+    const { agent, updates } = createMockAgentWithCapture();
+    // Top-level assistant on Opus, then subagent on Haiku (parent_tool_use_id set)
+    injectSession(agent, [
+      createAssistantMessage({ model: "claude-opus-4-20250514" }),
+      {
+        type: "assistant" as const,
+        parent_tool_use_id: "tool_use_123",
+        uuid: randomUUID(),
+        session_id: "test-session",
+        message: {
+          model: "claude-haiku-4-5-20251001",
+          content: [{ type: "text", text: "subagent response" }],
+          usage: {
+            input_tokens: 50,
+            output_tokens: 25,
+            cache_read_input_tokens: 0,
+            cache_creation_input_tokens: 0,
+          },
+        },
+      },
+      createResultMessageWithModel({
+        modelUsage: {
+          "claude-opus-4-20250514": {
+            inputTokens: 100,
+            outputTokens: 50,
+            cacheReadInputTokens: 20,
+            cacheCreationInputTokens: 10,
+            webSearchRequests: 0,
+            costUSD: 0.01,
+            contextWindow: 1000000,
+            maxOutputTokens: 16384,
+          },
+          "claude-haiku-4-5-20251001": {
+            inputTokens: 50,
+            outputTokens: 25,
+            cacheReadInputTokens: 0,
+            cacheCreationInputTokens: 0,
+            webSearchRequests: 0,
+            costUSD: 0.001,
+            contextWindow: 200000,
+            maxOutputTokens: 8192,
+          },
+        },
+      }),
+      { type: "system", subtype: "session_state_changed", state: "idle" },
+    ]);
+
+    await agent.prompt({ sessionId: "test-session", prompt: [{ type: "text", text: "test" }] });
+
+    const usageUpdate = updates.find((u: any) => u.update?.sessionUpdate === "usage_update");
+    expect(usageUpdate).toBeDefined();
+    // size should be 1000000 (Opus - the top-level model), NOT 200000 (Haiku subagent)
+    expect(usageUpdate.update.size).toBe(1000000);
+  });
+
+  it("prefix-matches when assistant model has date suffix but modelUsage key does not", async () => {
+    const { agent, updates } = createMockAgentWithCapture();
+    // The API response has the full versioned model ID on assistant messages,
+    // but the SDK's streaming path may key modelUsage by the shorter alias.
+    injectSession(agent, [
+      createAssistantMessage({ model: "claude-opus-4-6-20250514" }),
+      createResultMessageWithModel({
+        modelUsage: {
+          "claude-opus-4-6": {
+            inputTokens: 100,
+            outputTokens: 50,
+            cacheReadInputTokens: 20,
+            cacheCreationInputTokens: 10,
+            webSearchRequests: 0,
+            costUSD: 0.01,
+            contextWindow: 1000000,
+            maxOutputTokens: 16384,
+          },
+        },
+      }),
+      { type: "system", subtype: "session_state_changed", state: "idle" },
+    ]);
+
+    await agent.prompt({ sessionId: "test-session", prompt: [{ type: "text", text: "test" }] });
+
+    const usageUpdate = updates.find((u: any) => u.update?.sessionUpdate === "usage_update");
+    expect(usageUpdate).toBeDefined();
+    // Should match via prefix: "claude-opus-4-6-20250514".startsWith("claude-opus-4-6")
+    expect(usageUpdate.update.size).toBe(1000000);
+  });
+
+  it("prefix-matches when modelUsage key has date suffix but assistant model does not", async () => {
+    const { agent, updates } = createMockAgentWithCapture();
+    injectSession(agent, [
+      createAssistantMessage({ model: "claude-opus-4-6" }),
+      createResultMessageWithModel({
+        modelUsage: {
+          "claude-opus-4-6-20250514": {
+            inputTokens: 100,
+            outputTokens: 50,
+            cacheReadInputTokens: 20,
+            cacheCreationInputTokens: 10,
+            webSearchRequests: 0,
+            costUSD: 0.01,
+            contextWindow: 1000000,
+            maxOutputTokens: 16384,
+          },
+        },
+      }),
+      { type: "system", subtype: "session_state_changed", state: "idle" },
+    ]);
+
+    await agent.prompt({ sessionId: "test-session", prompt: [{ type: "text", text: "test" }] });
+
+    const usageUpdate = updates.find((u: any) => u.update?.sessionUpdate === "usage_update");
+    expect(usageUpdate).toBeDefined();
+    expect(usageUpdate.update.size).toBe(1000000);
+  });
+
+  it("synthetic assistant messages do not override lastAssistantModel", async () => {
+    const { agent, updates } = createMockAgentWithCapture();
+    // Real assistant on Opus, then a synthetic message (e.g. from /compact)
+    injectSession(agent, [
+      createAssistantMessage({ model: "claude-opus-4-20250514" }),
+      {
+        type: "assistant" as const,
+        parent_tool_use_id: null,
+        uuid: randomUUID(),
+        session_id: "test-session",
+        message: {
+          model: "<synthetic>",
+          content: [{ type: "text", text: "compacted" }],
+          usage: {
+            input_tokens: 0,
+            output_tokens: 0,
+            cache_read_input_tokens: 0,
+            cache_creation_input_tokens: 0,
+          },
+        },
+      },
+      createResultMessageWithModel({
+        modelUsage: {
+          "claude-opus-4-20250514": {
+            inputTokens: 100,
+            outputTokens: 50,
+            cacheReadInputTokens: 20,
+            cacheCreationInputTokens: 10,
+            webSearchRequests: 0,
+            costUSD: 0.01,
+            contextWindow: 1000000,
+            maxOutputTokens: 16384,
+          },
+        },
+      }),
+      { type: "system", subtype: "session_state_changed", state: "idle" },
+    ]);
+
+    await agent.prompt({ sessionId: "test-session", prompt: [{ type: "text", text: "test" }] });
+
+    const usageUpdate = updates.find((u: any) => u.update?.sessionUpdate === "usage_update");
+    expect(usageUpdate).toBeDefined();
+    // size should be 1000000 (Opus), not 200000 (the fallback if <synthetic> overrode the model)
+    expect(usageUpdate.update.size).toBe(1000000);
+  });
+});
--- a/src/tests/authorization.test.ts
+++ b/src/tests/authorization.test.ts
@@ -2,6 +2,20 @@ import { describe, expect, it, Mock, vi, afterEach, beforeEach } from "vitest";
 import { ClaudeAcpAgent } from "../acp-agent.js";
 import { AgentSideConnection } from "@agentclientprotocol/sdk";

+const mockQuery = vi.hoisted(() =>
+  vi.fn(() => ({
+    initializationResult: vi.fn().mockResolvedValue({
+      models: [{ value: "id", displayName: "name", description: "description" }],
+    }),
+    setModel: vi.fn(),
+    supportedCommands: vi.fn().mockResolvedValue([]),
+  })),
+);
+
+vi.mock("@anthropic-ai/claude-agent-sdk", () => ({
+  query: mockQuery,
+}));
+
 describe("authorization", () => {
  beforeEach(() => {
    vi.useFakeTimers();
@@ -17,20 +31,6 @@ describe("authorization", () => {
  });

  async function createAgentMock(): Promise<[ClaudeAcpAgent, Mock]> {
-    const mockQuery = vi.hoisted(() =>
-      vi.fn(() => ({
-        initializationResult: vi.fn().mockResolvedValue({
-          models: [{ value: "id", displayName: "name", description: "description" }],
-        }),
-        setModel: vi.fn(),
-        supportedCommands: vi.fn().mockResolvedValue([]),
-      })),
-    );
-
-    vi.mock("@anthropic-ai/claude-agent-sdk", () => ({
-      query: mockQuery,
-    }));
-
    const connectionMock = {
      sessionUpdate: async (_: any) => {},
    } as AgentSideConnection;
@@ -123,14 +123,17 @@ describe("authorization", () => {
      } as any,
    });
    expect(initializeResponse.authMethods).not.toContainEqual(
-      expect.objectContaining({ id: "claude-login" }),
+      expect.objectContaining({ id: "claude-ai-login" }),
+    );
+    expect(initializeResponse.authMethods).not.toContainEqual(
+      expect.objectContaining({ id: "console-login" }),
    );
    expect(initializeResponse.authMethods).toContainEqual(
      expect.objectContaining({ id: "gateway" }),
    );
  });

-  it("hide terminal auth even when terminal-auth is set", async () => {
+  it("hide claude auth but still show console login when terminal-auth is set", async () => {
    const [agent] = await createAgentMock();
    vi.stubGlobal("process", { ...process, argv: ["--hide-claude-auth"] });

@@ -140,7 +143,30 @@ describe("authorization", () => {
        _meta: { "terminal-auth": true },
      },
    });
-    expect(initializeResponse.authMethods).toStrictEqual([]);
+    expect(initializeResponse.authMethods).not.toContainEqual(
+      expect.objectContaining({ id: "claude-ai-login" }),
+    );
+    expect(initializeResponse.authMethods).toContainEqual(
+      expect.objectContaining({ id: "console-login" }),
+    );
+  });
+
+  it("hide claude auth but still show console login with terminal capability", async () => {
+    const [agent] = await createAgentMock();
+    vi.stubGlobal("process", { ...process, argv: ["--hide-claude-auth"] });
+
+    const initializeResponse = await agent.initialize({
+      protocolVersion: 1,
+      clientCapabilities: {
+        auth: { terminal: true },
+      },
+    });
+    expect(initializeResponse.authMethods).not.toContainEqual(
+      expect.objectContaining({ id: "claude-ai-login" }),
+    );
+    expect(initializeResponse.authMethods).toContainEqual(
+      expect.objectContaining({ id: "console-login" }),
+    );
  });

  it("show claude authentication", async () => {
@@ -152,7 +178,10 @@ describe("authorization", () => {
    });

    expect(initializeResponse.authMethods).toContainEqual(
-      expect.objectContaining({ id: "claude-login" }),
+      expect.objectContaining({ id: "claude-ai-login" }),
+    );
+    expect(initializeResponse.authMethods).toContainEqual(
+      expect.objectContaining({ id: "console-login" }),
    );
  });
 });