Merge pull request #1995 from deepseek-harness/worktree/remove-dsh-cli-demo

cleanup(cli): remove separate dsh-cli-demo app

docs/user/guide/index.i18n.yaml

@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write docs/user/guide/index.md
-index.md: ede09506a996193fe5cf4ae6cd9b64d3529798a6
-index.zh.md: 5f72a6d3099d2d4721eccebae92eacfe72d33bce
+index.md: a04698e29755d4a08b012f8b61accb79c470dcb0
+index.zh.md: 85990f49c7b3cdf7518237c6b558375b7a636243

docs/user/guide/index.md

@@ -12,12 +12,14 @@ Harness implements every capability an AI agent needs—including LLM calls, too
 # Select the LLM backend
 - name: '@deepseek-ai/dsh-llm-deepseek'
 
-# Select the one-shot application
-- id: cli-agent
-  name: '@deepseek-ai/dsh-cli-demo'
+# Compose one configured agent
+- id: agent-spine
+  name: '@deepseek-ai/dsh-agent-spine-demo'
   config:
-    provider: deepseek-official
-    model: deepseek-v4-flash
+    agents:
+      - id: main
+        provider: deepseek-official
+        model: deepseek-v4-flash
     workspaceContext: false
 ```
 

docs/user/guide/index.zh.md

@@ -12,12 +12,14 @@ Harness 将一个 AI Agent（智能体） 所需要的所有能力——LLM 调
 # Select the LLM backend
 - name: '@deepseek-ai/dsh-llm-deepseek'
 
-# Select the one-shot application
-- id: cli-agent
-  name: '@deepseek-ai/dsh-cli-demo'
+# Compose one configured agent
+- id: agent-spine
+  name: '@deepseek-ai/dsh-agent-spine-demo'
   config:
-    provider: deepseek-official
-    model: deepseek-v4-flash
+    agents:
+      - id: main
+        provider: deepseek-official
+        model: deepseek-v4-flash
     workspaceContext: false
 ```
 

docs/user/guide/quickstart.i18n.yaml

@@ -2,5 +2,5 @@
 # side as of the last confirmed-consistent state. Both languages carry equal authority;
 # after editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write docs/user/guide/quickstart.md
-quickstart.md: 8b84017ad33bf02579891bc4dcf83eaf7ec39022
-quickstart.zh.md: dfb24f8fa194866908406c709d059bf1f6595d59
+quickstart.md: 9d1a04f0a094919bb008a7e94564a6612e10732d
+quickstart.zh.md: 7182f63586f4f08524c643a7c6c1ad599b2de098

docs/user/guide/quickstart.md

@@ -22,6 +22,7 @@ pnpm -v
 git clone https://github.com/deepseek-ai/deepseek-harness-sdk.git
 cd deepseek-harness
 pnpm install
+pnpm run build
 ```
 
 Create the gitignored repository-root `.env`:
@@ -35,17 +36,16 @@ DEEPSEEK_API_KEY=sk-your-key-here
 Run a non-interactive task and print its final answer:
 
 ```sh
-pnpm run demo:headless "summarize the architecture of this workspace"
+pnpm run dsh run "summarize the architecture of this workspace"
 ```
 
-Headless runs one complete model/tool turn, persists the session, prints the result, and exits. Use `--output-format stream-json` when you need the canonical event stream.
+`dsh run` creates and persists a fresh session, prints the final assistant answer, and exits. While it runs, stderr prints the local browser URL where the session can be observed.
 
 ## Step 3: use the Web UI
 
-Build and start the browser interface:
+Start the browser interface:
 
 ```sh
-pnpm run build
 pnpm run dsh web
 ```
 
@@ -53,7 +53,7 @@ Open `http://127.0.0.1:3080`. The agent can read and write files, run commands,
 
 ## What happened
 
-headless-agent uses the `@deepseek-ai/dsh-cli-demo` app. `dsh web` instead boots the `web` profile: the [`dsh-base`](../../../packages/bundle/base/cordis.patch.yml) and [`dsh-web-app`](../../../packages/bundle/web-app/cordis.patch.yml) bundle patch layers composed over an empty root. Both select the DeepSeek model and capability plugins appropriate to their entry mode.
+`dsh run` boots the `headless` profile: [`dsh-base`](../../../packages/bundle/base/cordis.patch.yml), [`dsh-web-app`](../../../packages/bundle/web-app/cordis.patch.yml), and [`dsh-headless`](../../../packages/bundle/headless/cordis.patch.yml) compose over an empty root. `dsh web` uses the first two layers without the one-shot runner. Both select the DeepSeek model and capability plugins appropriate to their entry mode.
 
 ## Next steps
 

docs/user/guide/quickstart.zh.md

@@ -22,6 +22,7 @@ pnpm -v
 git clone https://github.com/deepseek-ai/deepseek-harness-sdk.git
 cd deepseek-harness
 pnpm install
+pnpm run build
 ```
 
 在仓库根目录创建已被 Git 忽略的 `.env`：
@@ -35,17 +36,16 @@ DEEPSEEK_API_KEY=sk-your-key-here
 运行一个非交互式任务并打印最终回答：
 
 ```sh
-pnpm run demo:headless "summarize the architecture of this workspace"
+pnpm run dsh run "summarize the architecture of this workspace"
 ```
 
-Headless 运行一个完整的模型/工具轮次，持久化会话，打印结果后退出。需要规范事件流时可使用 `--output-format stream-json`。
+`dsh run` 创建并持久化一个新会话，打印最终 assistant 回答，然后退出。运行期间，stderr 会打印可用于观察该会话的本地浏览器 URL。
 
 ## 第三步：使用 Web UI
 
-构建并启动浏览器界面：
+启动浏览器界面：
 
 ```sh
-pnpm run build
 pnpm run dsh web
 ```
 
@@ -53,7 +53,7 @@ pnpm run dsh web
 
 ## 回头看
 
-headless-agent 使用 `@deepseek-ai/dsh-cli-demo` app。`dsh web` 则启动 `web` profile：由 [`dsh-base`](../../../packages/bundle/base/cordis.patch.yml) 与 [`dsh-web-app`](../../../packages/bundle/web-app/cordis.patch.yml) 两个组合包的 patch 层在空根之上组合而成。二者都会根据各自入口模式选择 DeepSeek 模型和能力插件。
+`dsh run` 启动 `headless` profile：[`dsh-base`](../../../packages/bundle/base/cordis.patch.yml)、[`dsh-web-app`](../../../packages/bundle/web-app/cordis.patch.yml) 和 [`dsh-headless`](../../../packages/bundle/headless/cordis.patch.yml) 在空根之上组合。`dsh web` 使用前两层，不包含一次性 runner。二者都会根据各自入口模式选择 DeepSeek 模型和能力插件。
 
 ## 下一步