ci: use trusted publishing for TypeScript package (#486)

* ci: use trusted publishing for TypeScript package

* docs: scope npm announcement to DeepSeek Harness

.github/workflows/release-typescript.yml

@@ -1,6 +1,6 @@
 # Release checklist:
 # 1. Update packages/typescript/package.json and package-lock.json to the release version and merge them.
-# 2. Configure the NPM_TOKEN repository secret with publish access to the @agentscope-ai scope.
+# 2. Configure npm Trusted Publishing for agentscope-ai/ReMe and this workflow file.
 # 3. Run this workflow manually with the exact package version (an optional v prefix is accepted).
 # 4. Use the `next` tag for prereleases and `latest` only for stable releases.
 
@@ -128,6 +128,5 @@ jobs:
 
       - name: Publish to npm
         env:
-          NODE_AUTH_TOKEN: ${{ secrets.NPM_TOKEN }}
           NPM_TAG: ${{ inputs.npm_tag }}
         run: npm publish dist/typescript/*.tgz --access public --tag "${NPM_TAG}" --provenance

README.md

@@ -61,6 +61,8 @@ keeping the files under the user's control.
 
 ## 📰 News
 
+- [2026.08] - Published [`@agentscope-ai/reme`](https://www.npmjs.com/package/@agentscope-ai/reme), providing a native
+  ReMe memory integration for DeepSeek Harness.
 - [2026.08] - Published the [ReMe blog](https://agentscope-ai.github.io/ReMe/?doc=en-reme-blog), an end-to-end introduction to its local-first memory
   architecture, self-evolving workflows, hybrid search, proactive discovery, and benchmark results.
 - [2026.08] - [Experience-driven enhancement method](https://reme.agentscope.io/?doc=toolmemory-en) of agent tool-use execution built
@@ -98,6 +100,17 @@ cd ..
 
 The static build requires Node.js 22.13 or newer and makes Studio available from the source tree.
 
+### DeepSeek Harness Integration
+
+With the ReMe service running, install the npm package into the DeepSeek Harness Web profile:
+
+```bash
+dsh plugin --profile web add @agentscope-ai/reme
+```
+
+The plugin recalls relevant ReMe memory before agent steps and submits completed main-agent turns for automatic memory
+capture. See the [TypeScript integration guide](packages/typescript/README.md#deepseek-harness) for configuration.
+
 ### Environment Variables
 
 Configure environment variables when you want LLM-powered memory evolution or embedding retrieval. Embeddings are

README_ZH.md

@@ -53,6 +53,8 @@ Code 等 Agent 协作，在持续整理知识的同时，始终把文件控制
 
 ## 📰 新闻
 
+- [2026.08] - 发布 [`@agentscope-ai/reme`](https://www.npmjs.com/package/@agentscope-ai/reme)，为 DeepSeek Harness
+  提供原生 ReMe 记忆集成。
 - [2026.08] - 发布 [ReMe 博客](https://agentscope-ai.github.io/ReMe/?doc=zh-reme-blog)，系统介绍本地优先的记忆架构、自进化工作流、混合检索、
   主动发现与评测结果。
 - [2026.08] - 基于 ReMe 的智能体工具使用
@@ -91,6 +93,17 @@ cd ..
 
 静态构建要求 Node.js 22.13 或更高版本，并让源码安装可以直接使用 Studio。
 
+### DeepSeek Harness 集成
+
+启动 ReMe 服务后，将 npm 包安装到 DeepSeek Harness 的 Web profile：
+
+```bash
+dsh plugin --profile web add @agentscope-ai/reme
+```
+
+插件会在 Agent step 前检索相关 ReMe 记忆，并将主 Agent 已完成的对话提交给自动记忆任务。配置方法见
+[TypeScript 集成指南](packages/typescript/README_ZH.md#deepseek-harness)。
+
 ### 环境变量
 
 如果需要 LLM 驱动的记忆演化或 embedding 检索，可以配置环境变量。embedding 默认关闭，因此默认配置不会启动 embedding 模型，也不需要