diff --git a/README.md b/README.md index 2e8343a4..347ffc34 100644 --- a/README.md +++ b/README.md @@ -49,14 +49,25 @@ users retain control of the durable files. ## 📰 Latest Updates -- [2026.08] - Added independently installable ReMe memory plugins for DeepSeek Harness and OpenClaw. -- [2026.08] - Published the [ReMe blog](https://reme.agentscope.io/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/en/benchmarks/toolmemory) of agent tool-use execution built - on ReMe is available on [arXiv:2608.03403](https://arxiv.org/abs/2608.03403). -- [2026.07] - Introduced optional plugins: [Daily Paper](https://reme.agentscope.io/en/plugins/daily-paper) for paper discovery and - analysis, and [Auto Fin](https://reme.agentscope.io/en/plugins/auto-fin) for researching the latest 24 hours of topic-related CLS news - with local-memory search and validated historical wikilinks. +- [2026.09] - **[OpenClaw plugin](https://reme.agentscope.io/en/integrations/openclaw) released**: install it from + [ClawHub](https://clawhub.ai/agentscope-ai/plugins/reme-openclaw-plugin) or + [npm](https://www.npmjs.com/package/@agentscope-ai/reme-openclaw-plugin) to add native memory recall, automatic + conversation capture, and scheduled consolidation to OpenClaw. +- [2026.09] - **[DeepSeek Harness plugin](https://reme.agentscope.io/en/integrations/dsh) released**: install it from + [Awesome DSH Plugin](https://awesome-dsh-plugin.com/p/agentscope-ai/ReMe--integrations-dsh/) or + [npm](https://www.npmjs.com/package/@agentscope-ai/reme-dsh-plugin) for long-term-memory guidance, `reme_search`, + automatic memory, Auto Dream, and ReMe Status. +- [2026.08] - **ReMe blog published**: the [ReMe blog](https://reme.agentscope.io/en/reme-blog) introduces the + local-first memory architecture, self-evolving workflows, hybrid search, proactive discovery, and benchmark results. +- [2026.08] - **New ReMe ecosystem plugins**: [Daily Paper](https://reme.agentscope.io/en/plugins/daily-paper) + discovers and analyzes papers and generates file-native briefs, while + [Auto Fin](https://reme.agentscope.io/en/plugins/auto-fin) researches the latest 24 hours of topic-related CLS news + and builds traceable reports with local memory. Try them out. +- [2026.08] - **Plugin development support released**: use [Plugin Development](docs/en/plugin_development.md) and + [Plugin Management](docs/en/plugin_management.md) to extend ReMe with Components, Steps, and Jobs. Contributions and + new community plugins are welcome. +- [2026.08] - ReMe's [experience-driven enhancement method](https://reme.agentscope.io/en/benchmarks/toolmemory) for + agent tool use is available on [arXiv:2608.03403](https://arxiv.org/abs/2608.03403). - [2026.07] - Our paper [Remember Me, Refine Me: A Dynamic Procedural Memory Framework for Experience-Driven Agent Evolution](https://aclanthology.org/2026.findings-acl.829/) has been accepted to Findings of ACL 2026. @@ -180,10 +191,10 @@ lifecycle according to the capabilities of each runtime. | Agent | Recommended path | Available after integration | | ------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------- | -| **DeepSeek Harness** | Install [`@agentscope-ai/reme-dsh-plugin`](integrations/dsh/README.md) with `dsh plugin --profile web add @agentscope-ai/reme-dsh-plugin`. | Long-term memory guidance, the `reme_search` tool, and automatic capture of completed main-agent turns. | +| **DeepSeek Harness** | Install [`@agentscope-ai/reme-dsh-plugin`](integrations/dsh/README.md) with `dsh plugin --profile web add @agentscope-ai/reme-dsh-plugin`. | Configurable memory guidance, `reme_search`, automatic turn capture, scheduled Auto Dream, and ReMe Status. | | **OpenClaw** | Install [`@agentscope-ai/reme-openclaw-plugin`](integrations/openclaw/README.md) with `openclaw plugins install clawhub:@agentscope-ai/reme-openclaw-plugin`. | Native memory tools, recall before user-triggered runs, and automatic turn capture. | | **QwenPaw** | Embed ReMe in-process through its Python API. | Reuse the host lifecycle and model config while keeping memory local and file-based. | -| **Claude Code** | Start the streamable HTTP MCP service and install [the ReMe plugin](integrations/claude_code/reme). | MCP recall tools, the `reme-memory` skill, and a Stop hook that records sessions automatically. | +| **Claude Code** | Start the shared streamable HTTP MCP service and install [the ReMe plugin](integrations/claude_code/README.md). | Semantic, graph, and state recall through MCP, plus asynchronous session capture through a Stop hook. | | **Hermes** | Install [the ReMe provider](integrations/hermes_agent) and choose HTTP or embedded mode. | Recall before model calls and asynchronous `auto_memory` after each completed turn. | | **Codex and other CLI agents** | Install or copy the [ReMe Memory skill](skills/reme_memory/SKILL.md). | Search, read, and write memory through the CLI; automatic capture requires host lifecycle integration. | @@ -347,7 +358,8 @@ These guides cover the main user workflows and the runtime contracts implemented | [Proactive](docs/en/proactive.md) | Read interest topics safely and integrate them into a host agent's decision flow. | | [Application Scenarios](docs/en/reme_scene.md) | Follow concrete financial research, coding-memory, and personal knowledge-base examples. | | [Framework](docs/en/framework.md) | Understand Application, Job, Step, Component, service, configuration, and lifecycle boundaries. | -| [DSH plugin](integrations/dsh/README.md) and [OpenClaw plugin](integrations/openclaw/README.md) | Install native host adapters with independent dependencies and releases. | +| [Agent Integrations](docs/en/integrations.md) | Choose an interface and connect DSH, Claude Code, OpenClaw, Hermes, Codex, or another agent. | +| [DSH plugin](integrations/dsh/README.md) and [Claude Code plugin](integrations/claude_code/README.md) | Configure host-native recall, automatic capture, consolidation, and diagnostics. | | [CLI and Job API](docs/en/reference/cli.md) | Learn command syntax and use the generated default Job parameter reference. | | [Operations and Recovery](docs/en/operations.md) | Diagnose services, maintain indexes, and back up, migrate, or recover a workspace. | | [ReMe Blog](https://reme.agentscope.io/en/reme-blog) | Read the product story, design rationale, examples, and benchmark summary. | diff --git a/README_ZH.md b/README_ZH.md index 5ab57dad..0b34f8f3 100644 --- a/README_ZH.md +++ b/README_ZH.md @@ -47,15 +47,21 @@ ## 📰 最新动态 -- [2026.08] - 新增可独立安装的 DeepSeek Harness 和 OpenClaw ReMe 记忆插件。 -- [2026.08] - 发布 [ReMe 博客](https://reme.agentscope.io/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。 +- [2026.09] - **[OpenClaw 插件](https://reme.agentscope.io/zh/integrations/openclaw) 发布**：可通过 + [ClawHub](https://clawhub.ai/agentscope-ai/plugins/reme-openclaw-plugin) 或 + [npm](https://www.npmjs.com/package/@agentscope-ai/reme-openclaw-plugin) 安装，为 OpenClaw 提供原生记忆召回、自动对话捕获和定时整理能力。 +- [2026.09] - **[DeepSeek Harness 插件](https://reme.agentscope.io/zh/integrations/dsh) 发布**：可通过 + [Awesome DSH Plugin](https://awesome-dsh-plugin.com/p/agentscope-ai/ReMe--integrations-dsh/) 或 + [npm](https://www.npmjs.com/package/@agentscope-ai/reme-dsh-plugin) 安装，提供长期记忆指引、`reme_search`、自动记忆、Auto Dream 和 ReMe Status。 +- [2026.08] - **ReMe 博客发布**：[ReMe 博客](https://reme.agentscope.io/zh/reme-blog) 系统介绍了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] - **新增 ReMe 生态插件**：[每日论文](https://reme.agentscope.io/zh/plugins/daily-paper) 可自动发现、解析论文并生成文件化简报； + [Auto Fin](https://reme.agentscope.io/zh/plugins/auto-fin) 可研究最近 24 小时的主题相关财联社新闻，并结合本地记忆构建可追溯报告。欢迎体验。 +- [2026.08] - **插件开发能力上线**：参考 [插件开发](docs/zh/plugin_development.md) 与 [插件管理](docs/zh/plugin_management.md)， + 为 ReMe 扩展 Component、Step 和 Job；欢迎开发并分享你的插件。 - [2026.08] - 基于 ReMe 的智能体工具使用 - [经验驱动增强方法](https://reme.agentscope.io/zh/benchmarks/toolmemory)已发布，见 + [经验驱动增强方法](https://reme.agentscope.io/zh/benchmarks/toolmemory) 已发布，见 [arXiv:2608.03403](https://arxiv.org/abs/2608.03403)。 -- [2026.07] - 新增可选插件：[每日论文](https://reme.agentscope.io/zh/plugins/daily-paper)用于论文发现与解析， - [Auto Fin](https://reme.agentscope.io/zh/plugins/auto-fin)用于研究最近 24 小时的主题相关财联社新闻，通过本地记忆搜索回顾历史材料并构建 - wikilink。 - [2026.07] - 我们的论文 [Remember Me, Refine Me: A Dynamic Procedural Memory Framework for Experience-Driven Agent Evolution](https://aclanthology.org/2026.findings-acl.829/) 已被 Findings of ACL 2026 接收。 @@ -178,10 +184,10 @@ runtime 的能力，将记忆指引、召回和捕获接入 Agent 生命周期 | Agent | 推荐接入方式 | 接入后能力 | | -------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------- | -| **DeepSeek Harness** | 使用 `dsh plugin --profile web add @agentscope-ai/reme-dsh-plugin` 安装 [`@agentscope-ai/reme-dsh-plugin`](integrations/dsh/README_ZH.md)。 | 长期记忆指引、`reme_search` 工具，以及自动捕获已完成的主 Agent 对话。 | +| **DeepSeek Harness** | 使用 `dsh plugin --profile web add @agentscope-ai/reme-dsh-plugin` 安装 [`@agentscope-ai/reme-dsh-plugin`](integrations/dsh/README_ZH.md)。 | 可配置记忆指引、`reme_search`、自动对话捕获、定时 Auto Dream 和 ReMe Status。 | | **OpenClaw** | 使用 `openclaw plugins install clawhub:@agentscope-ai/reme-openclaw-plugin` 安装 [`@agentscope-ai/reme-openclaw-plugin`](integrations/openclaw/README_ZH.md)。 | 原生记忆工具、用户触发运行前召回和自动对话捕获。 | | **QwenPaw** | 通过 Python API 在进程内嵌入 ReMe。 | 复用宿主生命周期和模型配置，同时保持记忆本地、文件化。 | -| **Claude Code** | 启动 streamable HTTP MCP service，并安装 [ReMe 插件](integrations/claude_code/reme)。 | MCP 召回工具、`reme-memory` skill，以及自动记录会话的 Stop hook。 | +| **Claude Code** | 启动共享的 streamable HTTP MCP service，并安装 [ReMe 插件](integrations/claude_code/README.md)。 | 通过 MCP 进行语义、图关系和状态召回，并由 Stop Hook 异步捕获会话。 | | **Hermes** | 安装 [ReMe provider](integrations/hermes_agent)，并选择 HTTP 或 Embedded 模式。 | 模型调用前召回，每轮对话完成后异步执行 `auto_memory`。 | | **Codex 及其他 CLI Agent** | 安装或复制 [ReMe Memory skill](skills/reme_memory/SKILL.md)。 | 通过 CLI 搜索、读取和写入记忆；自动捕获需要显式接入宿主生命周期。 | @@ -337,7 +343,8 @@ ReMe 通过 Agent 多轮搜索与读取的方式，评测多会话和超长上 | [Proactive](docs/zh/proactive.md) | 安全读取兴趣主题，并将其接入宿主 Agent 的决策流程。 | | [应用场景](docs/zh/reme_scene.md) | 查看金融研究、研发记忆和个人知识库的完整使用示例。 | | [框架说明](docs/zh/framework.md) | 理解 Application、Job、Step、Component、service、配置和生命周期边界。 | -| [DSH 插件](integrations/dsh/README_ZH.md) 与 [OpenClaw 插件](integrations/openclaw/README_ZH.md) | 安装具有独立依赖和发布周期的原生宿主适配器。 | +| [Agent 集成](docs/zh/integrations.md) | 选择接口，并将 DSH、Claude Code、OpenClaw、Hermes、Codex 或其他 Agent 接入 ReMe。 | +| [DSH 插件](integrations/dsh/README_ZH.md) 与 [Claude Code 插件](integrations/claude_code/README.md) | 配置宿主原生召回、自动捕获、记忆整理与诊断。 | | [CLI 与 Job API](docs/zh/reference/cli.md) | 查询命令语法，以及由默认配置自动生成的 Job 参数参考。 | | [运维与恢复](docs/zh/operations.md) | 诊断服务、维护索引，并备份、迁移和恢复 workspace。 | | [ReMe 博客](https://reme.agentscope.io/zh/reme-blog) | 了解完整产品故事、设计动机、使用示例和评测摘要。 | diff --git a/docs/en/integrations.md b/docs/en/integrations.md index 8902b2e2..e93f8358 100644 --- a/docs/en/integrations.md +++ b/docs/en/integrations.md @@ -14,9 +14,9 @@ ReMe keeps memory in an independent service and a user-owned workspace. Multiple | Local script or hook | ReMe CLI | | Application backend | HTTP Client | | Tool-protocol host | MCP | -| DeepSeek Harness | `@agentscope-ai/reme-dsh-plugin` | -| OpenClaw | `@agentscope-ai/reme-openclaw-plugin` | -| Claude Code | MCP + Skill + Stop Hook | +| DeepSeek Harness | [`@agentscope-ai/reme-dsh-plugin`](./integrations/dsh.md) profile bundle | +| OpenClaw | [`@agentscope-ai/reme-openclaw-plugin`](./integrations/openclaw.md) | +| Claude Code | [Shared HTTP MCP + Skill + Stop Hook](./integrations/claude-code.md) | | Hermes Agent | Memory provider adapter | | Codex or another coding agent | `reme_memory` Skill or MCP | @@ -43,14 +43,48 @@ Use `service.jobs` to expose a read-only subset or keep write tools in a separat It deliberately avoids silently modifying Python environments, stopping unknown processes on port conflicts, writing recalled tool output back as conversation source, or persisting credentials. -## TypeScript, OpenClaw, and DeepSeek Harness +## DeepSeek Harness -Install the self-contained [DeepSeek Harness](./integrations/dsh.md) or [OpenClaw](./integrations/openclaw.md) plugin. -Each package owns its ReMe HTTP boundary and can evolve with its host independently. +Install the self-contained [DeepSeek Harness plugin](./integrations/dsh.md): + +```bash +dsh plugin --profile web add @agentscope-ai/reme-dsh-plugin +``` + +Release links: [Awesome DSH Plugin](https://awesome-dsh-plugin.com/p/agentscope-ai/ReMe--integrations-dsh/) and +[npm](https://www.npmjs.com/package/@agentscope-ai/reme-dsh-plugin). + +It injects long-term-memory usage guidance into new root-agent sessions and exposes the read-only `reme_search` tool; +it does not preload the full memory history into the prompt. Completed user/assistant turns can be submitted to +`auto_memory` in background batches, while a timezone-aware schedule runs `auto_dream` to consolidate daily notes. + +DSH settings configure the endpoint, guidance language, search limits, capture interval, root-agent filtering, and +consolidation schedule. The ReMe Status page exposes Overview, Auto Memory, Memory Consolidation, Components, Journal, +and Personal Knowledge Base views. Runtime counters are diagnostic state; workspace Markdown remains the durable source +of truth. + +## OpenClaw + +Install the independently published [OpenClaw plugin](./integrations/openclaw.md): + +```bash +openclaw plugins install clawhub:@agentscope-ai/reme-openclaw-plugin +``` + +Release links: [ClawHub](https://clawhub.ai/agentscope-ai/plugins/reme-openclaw-plugin) and +[npm](https://www.npmjs.com/package/@agentscope-ai/reme-openclaw-plugin). The plugin provides its own host-specific +ReMe HTTP boundary and release lifecycle. ## Claude Code -`integrations/claude_code/` provides streamable HTTP MCP configuration, a `reme-memory` Skill, and a Stop hook that calls `auto_memory_cc`. Follow that directory's README for installation. +The [Claude Code plugin](./integrations/claude-code.md) connects every Claude Code window to one ReMe HTTP process at +`http://127.0.0.1:2333/mcp` by default. The `reme-memory` Skill selects among semantic `search`, topological `traverse`, +and state-oriented `daily_list` / `frontmatter_read`, then reads and cites the relevant workspace paths. + +On Stop, the hook passes only the Claude Code `session_id` to the server-side `auto_memory_cc` job. On POSIX systems it +detaches the potentially long model call so Claude Code can stop immediately; unreachable-service and other best-effort +failures are written to the plugin log instead of blocking the host. ReMe resolves the local transcript, and repeated +Stop events with no new messages do not create duplicate memory. ## Hermes Agent diff --git a/docs/en/reme-blog.md b/docs/en/reme-blog.md index c39f0c91..1e7f3cf1 100644 --- a/docs/en/reme-blog.md +++ b/docs/en/reme-blog.md @@ -316,10 +316,10 @@ that best fits their runtime environment and share the same local memory workspa | Agent | Recommended integration | Capabilities after integration | |-------|-------------------------|--------------------------------| -| **DeepSeek Harness** | Install [`@agentscope-ai/reme-dsh-plugin`](../../integrations/dsh/README.md) as a DSH profile bundle. | Long-term memory guidance, `reme_search`, automatic capture of completed main-agent turns, and scheduled Auto Dream. | -| **OpenClaw** | Install [`@agentscope-ai/reme-openclaw-plugin`](../../integrations/openclaw/README.md) as the native memory plugin. | Recall before conversational root-agent runs, explicit search, automatic turn capture, and scheduled Auto Dream. | +| **DeepSeek Harness** | Install [`@agentscope-ai/reme-dsh-plugin`](https://reme.agentscope.io/en/integrations/dsh) as a DSH profile bundle. | Long-term memory guidance, `reme_search`, automatic capture of completed main-agent turns, scheduled Auto Dream, and ReMe Status. | +| **OpenClaw** | Install [`@agentscope-ai/reme-openclaw-plugin`](https://reme.agentscope.io/en/integrations/openclaw) as the native memory plugin. | Recall before conversational root-agent runs, explicit search, automatic turn capture, scheduled Auto Dream, and status diagnostics. | | **QwenPaw** | Embed ReMe in-process through the Python API. | Reuse the host application's lifecycle and model configuration while keeping memories local and file-based. | -| **Claude Code** | Start the streamable HTTP MCP Service and install [`integrations/claude_code/reme`](../../integrations/claude_code/reme). | MCP memory-recall tools, the `reme-memory` skill, and a Stop hook that automatically records sessions. | +| **Claude Code** | Start the shared streamable HTTP MCP Service and install the [Claude Code plugin](https://reme.agentscope.io/en/integrations/claude-code). | Semantic, graph, and state recall through MCP, plus asynchronous session capture through a Stop hook. | | **Hermes** | Install [`integrations/hermes_agent`](../../integrations/hermes_agent) and choose HTTP or embedded mode. | Automatically recall relevant memories before model calls and invoke `auto_memory` asynchronously after each conversation turn. | | **Codex and other CLI-capable agents** | Copy or install [`skills/reme_memory/SKILL.md`](../../skills/reme_memory/SKILL.md). | Search, read, and write memories through the CLI; automatic recording requires the host agent to integrate explicitly with the conversation lifecycle. | diff --git a/docs/zh/integrations.md b/docs/zh/integrations.md index 0e9901be..a1d9152c 100644 --- a/docs/zh/integrations.md +++ b/docs/zh/integrations.md @@ -14,9 +14,9 @@ ReMe 把记忆能力放在独立服务和用户拥有的 workspace 中。Agent | 本机脚本或 Hook | ReMe CLI | | 应用后端 | HTTP Client | | 支持工具协议的 Agent | MCP | -| DeepSeek Harness | `@agentscope-ai/reme-dsh-plugin` | -| OpenClaw | `@agentscope-ai/reme-openclaw-plugin` | -| Claude Code | MCP + Skill + Stop Hook | +| DeepSeek Harness | [`@agentscope-ai/reme-dsh-plugin`](./integrations/dsh.md) profile bundle | +| OpenClaw | [`@agentscope-ai/reme-openclaw-plugin`](./integrations/openclaw.md) | +| Claude Code | [共享 HTTP MCP + Skill + Stop Hook](./integrations/claude-code.md) | | Hermes Agent | Memory provider adapter | | Codex 或其他 coding agent | `reme_memory` Skill 或 MCP | @@ -56,20 +56,44 @@ Skill 不应： - 把召回的工具结果再次写入对话来源； - 将密钥或敏感信息写入记忆。 -## TypeScript、OpenClaw 与 DeepSeek Harness +## DeepSeek Harness -安装自包含、独立发布的 [DeepSeek Harness 插件](./integrations/dsh.md)或 -[OpenClaw 插件](./integrations/openclaw.md)。每个包拥有自己的 ReMe HTTP 边界，可以跟随对应宿主独立演进。 +安装自包含的 [DeepSeek Harness 插件](./integrations/dsh.md)： + +```bash +dsh plugin --profile web add @agentscope-ai/reme-dsh-plugin +``` + +发布页：[Awesome DSH Plugin](https://awesome-dsh-plugin.com/p/agentscope-ai/ReMe--integrations-dsh/) 和 +[npm](https://www.npmjs.com/package/@agentscope-ai/reme-dsh-plugin)。 + +插件会在新的根 Agent 会话中注入长期记忆使用指引，并提供只读 `reme_search` 工具；它不会把所有历史记忆预先塞入上下文。 +已完成的用户/助手对话可以分批在后台交给 `auto_memory`，并由带时区的计划任务调用 `auto_dream` 整理 daily note。 + +DSH 设置可配置服务地址、指引语言、搜索数量、捕获间隔、根 Agent 过滤和整理计划。ReMe Status 页面包含 Overview、 +Auto Memory、Memory Consolidation、Components、Journal 和 Personal Knowledge Base 六个视图。运行时计数只用于诊断， +workspace 中的 Markdown 仍是持久事实来源。 + +## OpenClaw + +安装独立发布的 [OpenClaw 插件](./integrations/openclaw.md)： + +```bash +openclaw plugins install clawhub:@agentscope-ai/reme-openclaw-plugin +``` + +发布页：[ClawHub](https://clawhub.ai/agentscope-ai/plugins/reme-openclaw-plugin) 和 +[npm](https://www.npmjs.com/package/@agentscope-ai/reme-openclaw-plugin)。该插件拥有自己的宿主适配 ReMe HTTP 边界和发布周期。 ## Claude Code -仓库的 `integrations/claude_code/` 提供： +[Claude Code 插件](./integrations/claude-code.md) 默认让所有 Claude Code 窗口连接同一个 +`http://127.0.0.1:2333/mcp` ReMe HTTP 进程。`reme-memory` Skill 会在语义 `search`、图关系 `traverse` 和状态查询 +`daily_list` / `frontmatter_read` 之间选择，再读取并引用相关 workspace 路径。 -- streamable HTTP MCP 配置； -- `reme-memory` Skill； -- 会话停止时调用 `auto_memory_cc` 的 Hook。 - -完整安装步骤以仓库中的 `integrations/claude_code/README.md` 为准。 +会话 Stop 时，Hook 只把 Claude Code `session_id` 交给服务端 `auto_memory_cc` Job。在 POSIX 系统上，它会脱离可能耗时的 +模型调用，让 Claude Code 立即停止；服务不可达等 best-effort 失败只写入插件日志，不阻塞宿主。ReMe 会解析本地 transcript； +重复 Stop 且没有新消息时，不会重复生成记忆。 ## Hermes Agent diff --git a/docs/zh/integrations/claude-code.md b/docs/zh/integrations/claude-code.md index 17f78ab4..ee3f1703 100644 --- a/docs/zh/integrations/claude-code.md +++ b/docs/zh/integrations/claude-code.md @@ -5,14 +5,17 @@ description: 通过 MCP、reme-memory Skill 和 Stop Hook 将 Claude Code 连接 # Claude Code 集成 -ReMe 的 Claude Code 插件提供长期记忆召回，并在每次会话结束后异步记录对话。Daily 到 digest 的整理仍由共享的 ReMe 服务负责。 +ReMe 的 Claude Code 插件负责召回和会话捕获：所有 Claude Code 窗口共享一个 ReMe HTTP MCP 服务，Daily 到 digest 的 +整理、watcher 和 dream cron 也只在这个服务中运行一份。 ## 能力 -- 通过 MCP 使用 `search`、`traverse`、`daily_list`、`frontmatter_read`、`read`、`auto_memory_cc` 等工具； -- `reme-memory` Skill 在回答前召回长期记忆并保留来源路径； +- 通过 MCP 使用 `search`、`traverse`、`daily_list`、`frontmatter_read`、`read`、`version`、`health_check` 和 + `auto_memory_cc` 等工具； +- `reme-memory` Skill 区分语义、图关系和状态三种查询，再用 `read` 读取命中内容并保留 workspace-relative 来源路径； +- Skill 可通过 `version` 和 `health_check` 检查共享服务；工具缺失时应提示启动 ReMe，不得猜测历史记忆； - Stop Hook 只把 Claude Code `session_id` 传给服务端，服务端从本地 transcript 解析会话； -- 记录在脱离 Claude Code 的后台进程中进行，不延迟退出；服务不可用时记录日志并结束。 +- POSIX 系统上的记录会在脱离 Claude Code 的后台进程中进行，不延迟退出；服务不可用时记录日志并结束。 ## 部署模型 @@ -34,6 +37,13 @@ reme start service.backend=http 默认搜索使用 BM25；只有启用向量检索时才需要 Embedding 配置。 +请勿将默认 HTTP 服务直接暴露到不可信网络；跨主机使用时应在反向代理层提供认证和 TLS，并限制可调用 Job。 + +自动捕获还要求 ReMe 服务进程能读取 Claude Code transcript。它默认在服务端的 `~/.claude/projects` +下查找会话；因此 ReMe 应与 Claude Code 运行在同一主机，或者将 transcript 挂载/同步到服务端，并在启动 +ReMe 时用 `CLAUDE_CONFIG_DIR` 指向对应目录。如果服务端无法访问 transcript，远程 MCP 召回等功能仍可使用，但 +`auto_memory_cc` 会因没有消息而跳过，不会生成记忆。 + ## 安装插件 在 Claude Code 中运行： @@ -51,11 +61,14 @@ reme start service.backend=http - 自动记忆 Hook：`integrations/claude_code/reme/hooks/auto_memory.py`； - Hook 日志：`integrations/claude_code/reme/logs/auto_memory_hook.log`； - 默认 transcript 根目录：`~/.claude/projects`； -- 可通过 `CLAUDE_CONFIG_DIR` 修改 transcript 根目录； +- 启动 ReMe 服务时，可通过 `CLAUDE_CONFIG_DIR` 修改服务端的 transcript 根目录； - 可通过 `REME_HOST`、`REME_PORT` 覆盖 Hook 使用的服务地址。 Hook 需要 `python3` 位于 `PATH`。MCP 工具名前缀可能随 Claude Code 版本包含 server segment；Skill 使用 `mcp__reme__*` 匹配这一差异。 +Hook 是 best-effort 的：它不会因记忆服务故障而阻塞 Claude Code。服务端会保留 transcript 处理进度；同一会话重复触发 Stop 且没有新消息时， +`auto_memory_cc` 会跳过重复的记忆生成和打标。 + ## 验证 1. `reme health_check` 返回健康； diff --git a/docs/zh/reme-blog.md b/docs/zh/reme-blog.md index e7309b07..0c4c3826 100644 --- a/docs/zh/reme-blog.md +++ b/docs/zh/reme-blog.md @@ -334,10 +334,10 @@ ReMe 既可以作为本地记忆服务，通过 CLI、HTTP API 或 MCP Server | Agent | 推荐接入方式 | 接入后能力 | |----------------------------------------|--------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------| -| **DeepSeek Harness** | 将 [`@agentscope-ai/reme-dsh-plugin`](../../integrations/dsh/README_ZH.md) 安装为 DSH profile bundle。 | 长期记忆指引、`reme_search`、自动捕获主 Agent 已完成的对话，以及定时 Auto Dream。 | -| **OpenClaw** | 将 [`@agentscope-ai/reme-openclaw-plugin`](../../integrations/openclaw/README_ZH.md) 安装为原生 memory plugin。 | 根 Agent 对话运行前召回、显式搜索、自动捕获对话，以及定时 Auto Dream。 | +| **DeepSeek Harness** | 将 [`@agentscope-ai/reme-dsh-plugin`](https://reme.agentscope.io/zh/integrations/dsh) 安装为 DSH profile bundle。 | 长期记忆指引、`reme_search`、自动捕获主 Agent 已完成的对话、定时 Auto Dream 和 ReMe Status。 | +| **OpenClaw** | 将 [`@agentscope-ai/reme-openclaw-plugin`](https://reme.agentscope.io/zh/integrations/openclaw) 安装为原生 memory plugin。 | 根 Agent 对话运行前召回、显式搜索、自动捕获对话、定时 Auto Dream 和状态诊断。 | | **QwenPaw** | 通过 Python API 在进程内嵌入 ReMe。 | 复用宿主应用的生命周期和模型配置，同时保持记忆本地、文件化。 | -| **Claude Code** | 启动 streamable HTTP MCP Service，并安装 [`integrations/claude_code/reme`](../../integrations/claude_code/reme)。 | MCP 记忆召回工具、`reme-memory` skill，以及自动记录会话的 Stop hook。 | +| **Claude Code** | 启动共享 streamable HTTP MCP Service，并安装 [Claude Code 插件](https://reme.agentscope.io/zh/integrations/claude-code)。 | 通过 MCP 进行语义、图关系和状态召回，并由 Stop Hook 异步捕获会话。 | | **Hermes** | 安装 [`integrations/hermes_agent`](../../integrations/hermes_agent)，并选择 HTTP 或 Embedded 模式。 | 在模型调用前自动召回相关记忆，并在每轮对话完成后异步调用 `auto_memory`。 | | **Codex 等支持 CLI 的 Agent** | 复制或安装 [`skills/reme_memory/SKILL.md`](../../skills/reme_memory/SKILL.md)。 | 通过 CLI 搜索、读取和写入记忆；自动记录需要宿主 Agent 显式接入会话生命周期。 | diff --git a/integrations/dsh/README.md b/integrations/dsh/README.md index 11a5d6da..e50d3973 100644 --- a/integrations/dsh/README.md +++ b/integrations/dsh/README.md @@ -48,7 +48,9 @@ For development and screenshots, use an isolated directory outside the repositor ### 3.2 Install the DSH bundle -Install the published package: +Install the published package. See its listings on +[Awesome DSH Plugin](https://awesome-dsh-plugin.com/p/agentscope-ai/ReMe--integrations-dsh/) and +[npm](https://www.npmjs.com/package/@agentscope-ai/reme-dsh-plugin): ```bash dsh plugin --profile web add @agentscope-ai/reme-dsh-plugin diff --git a/integrations/dsh/README_ZH.md b/integrations/dsh/README_ZH.md index 56329437..6dc635ce 100644 --- a/integrations/dsh/README_ZH.md +++ b/integrations/dsh/README_ZH.md @@ -47,7 +47,9 @@ reme start workspace_dir=/absolute/path/to/your/reme-workspace \ ### 3.2 安装插件 -安装已发布版本： +安装已发布版本，发布信息见 +[Awesome DSH Plugin](https://awesome-dsh-plugin.com/p/agentscope-ai/ReMe--integrations-dsh/) 和 +[npm](https://www.npmjs.com/package/@agentscope-ai/reme-dsh-plugin)： ```bash dsh plugin --profile web add @agentscope-ai/reme-dsh-plugin diff --git a/integrations/openclaw/README.md b/integrations/openclaw/README.md index f87a5051..ecf2640e 100644 --- a/integrations/openclaw/README.md +++ b/integrations/openclaw/README.md @@ -54,7 +54,9 @@ ReMe HTTP does not add API-key authentication. Keep it on loopback or place it b ## 2. Install the plugin -From a released package: +From a released package. See its listings on +[ClawHub](https://clawhub.ai/agentscope-ai/plugins/reme-openclaw-plugin) and +[npm](https://www.npmjs.com/package/@agentscope-ai/reme-openclaw-plugin): ```bash openclaw plugins install clawhub:@agentscope-ai/reme-openclaw-plugin diff --git a/integrations/openclaw/README_ZH.md b/integrations/openclaw/README_ZH.md index 907c3475..cdc1eae3 100644 --- a/integrations/openclaw/README_ZH.md +++ b/integrations/openclaw/README_ZH.md @@ -51,7 +51,9 @@ ReMe HTTP 本身不增加 API Key 认证。应保持 loopback，或在可信的 ## 2. 安装插件 -安装发布包： +安装发布包，发布信息见 +[ClawHub](https://clawhub.ai/agentscope-ai/plugins/reme-openclaw-plugin) 和 +[npm](https://www.npmjs.com/package/@agentscope-ai/reme-openclaw-plugin)： ```bash openclaw plugins install clawhub:@agentscope-ai/reme-openclaw-plugin