MemoryBridge：为 AI Agent 提供本地持久化记忆与跨 Session 恢复
MemoryBridge（@user_6c78b4cc/memory-bridge）是 SkillHub 社区 ai-agent 类 Skill，面向「新对话丢上下文」痛点，用 memory_cli.py 十条本地命令完成写日志、合并 MEMORY.md、中文搜索与 bridge 恢复。无需 API Key，配合 SKILL.md 规定 Agent 落盘与新 Session 主动桥接。文章说明 SkillHub 安装路径、install.sh 依赖与典型命令，并提示 OpenClaw 默认目录与 Cursor 等环境的 workdir 对齐事项。
Read Moredsh-mate-companion: Packaging Persisto Mate as a DeepSeek Harness Plugin
`m-rui001/dsh-mate-companion` is an AI companion plugin for DeepSeek Harness (DSH), designed to add persistent memory, emotion simulation, and body clock capabilities to agents. The plugin is MIT-licensed and provides core features such as system prompt snippets, a kernel for processing emotional states and offline credits, and memory retrieval and consolidation. It includes three tools—memory, thinking, and debugging—and supports multilingual switching. Developers should note that it is currently in preview and has limitations, including self-reporting bias.
Read Moredsh-memory: Persistent Memory Plugin for DeepSeek Harness
The `dsh-memory` plugin for DeepSeek Harness (DSH) is designed to address the loss of context after an agent session restart. It implements persistent storage using pure Markdown files and supports user-level and project-level memory isolation, ensuring agent consistency across different sessions. Its core features include cross-session memory recall, a visual memory manager, and project memory isolation. When using it, note the ordering of the `--profile` parameter and restart the service to register the tool. The plugin has zero dependencies and ensures safety through atomic writes and boundary design, making it well suited for agent development scenarios requiring long-term context management.
Read Moredsh-honcho: Providing Persistent Memory for DeepSeek Harness
dsh-honcho is a DeepSeek Harness plugin maintained by Plastic Labs that provides persistent memory through Honcho, addressing context loss issues. It supports injecting context at session startup, background message capture, and built-in tools, and offers status viewing. Installation requires a Node.js environment, adding the plugin via commands, and setting the API key, companion name, and a shared workspace value in the configuration file to enable memory sharing.
Read Moredsh-memory: A Persistent Atomic Fact Memory Plugin for DeepSeek Harness
The `dsh-memory` plugin for DeepSeek Harness (DSH) is designed to address the loss of context in multi-turn agent conversations. Built on Cordis, it uses local JSON documents for persistent storage and does not rely on external databases. Its core is the atomic fact model (Subject-Predicate-Object), which uniformly manages semantic, episodic, and procedural memory. The plugin provides tools such as `recall`, `remember`, and `forget`, supports capturing user preferences through trigger phrases, and includes PII privacy protection mechanisms. It also supports `user.md` file synchronization and dynamic context injection, facilitating manual intervention and system reconstruction logging. Although it currently only supports BM25 semantic retrieval and LLM extraction is disabled by default,
Read Moredsh-maestro-memory: Persistent Memory and Todo Plugin for DeepSeek Harness
`dsh-maestro-memory` is a DeepSeek Harness (DSH) plugin designed to address context loss caused by session termination. It enables cross-session memory and to-do list management by persisting data in `~/.dsh/memories`. The plugin supports multi-track memory (global, user, etc.), to-do list management, and filtering of sensitive information. Installation requires Node.js 22+, pnpm 11+, and the DSH master version. It also provides health checks, memory repair, and data synchronization, making it suitable for scenarios that require long-term memory tracking.
Read Moredsh-agent-team: A Persistent Multi-Agent Team Collaboration Solution for DeepSeek Harness
`dsh-agent-team` is a persistent collaboration plugin for DeepSeek Harness, maintained by wowyuarm. It allows agents to retain memory after session interruptions or restarts, enabling long-term collaboration and independent management of multiple agents through workspaces, channels, and task threads. Users can manage members, agents, and permissions via the Web UI. When installing, note that it must be launched with the compiled CLI and used in a trusted environment to avoid permission risks.
Read MoreDeepSeek Harness Persistent Memory Plugin: dsh-memory-panel
The `dsh-memory-panel` plugin provides persistent, searchable conversation memory management for DeepSeek Harness (DSH), addressing memory loss caused by process restarts. Its core features include memory tools (save, search, list, retrieve, and delete), highlighted memories automatically injected into system prompts, and a visual management panel with a JSON API. Data is stored in the `.dsh-memory` folder in the user's home directory, with concurrent write protection, weighted score-based sorting (title > tags > content), and Chinese word-segmentation search.
Read MoreDeepSeek Harness Plugin: Persistent Five-Layer Memory System (nattocb/dsh-plugin-memory)
`nattocb/dsh-plugin-memory` provides DeepSeek Harness (DSH) with a persistent five-layer memory model across sessions. The plugin injects memory through the Cordis interface and registers six dedicated tools, structuring memory into five layers: user identity, profile, project context (with indexing and topic splitting), daily logs, and skills. It supports relevance-based injection, keyword scoring, and automatic LLM extraction, enabling agents to retrieve and update information across sessions. The plugin has no LLM dependency and runs entirely via DSH's internal interfaces, aiming to address the problem of agents losing context within a single session.
Read Moredsh-memory: A Codex-Style Persistent Memory Plugin for DeepSeek Harness
The DeepSeek Harness plugin `dsh-memory` provides persistent memory for agents, addressing the loss of context in multi-turn conversations. Its core features include persistent storage, automatic summary injection, a memory toolset, and configuration via a web interface. Installation requires specifying a version hash and restarting the environment. The plugin operates in the `$DSH_HOME/memories/` directory and achieves long-term memory capabilities similar to Codex by automatically distilling conversation turns and periodically merging them into a global summary. It supports initializing seed memory from `AGENTS.md` and provides tools for reading and writing memory.
Read MoreIntroduction to the DeepSeek Harness (DSH) Persistent Memory Plugin
The `disc0nct/dsh-memory-plugin` plugin for DeepSeek Harness (DSH) addresses the challenge of context loss caused by session isolation in agent development. It provides persistent storage, hybrid retrieval, and categorized management, supporting the invocation of seven tools (such as storage, search, and statistics) via `ctx.tools` to manage cross-session memory. The plugin includes atomic writes, concurrency safety, and data recovery mechanisms, supports up to 5,000 memories, and effectively improves the coherence of long conversations.
Read Moredsh-memory: DeepSeek Harness Persistent Memory System
dsh-memory is a memory system plugin for DeepSeek Harness (DSH), designed to persist agent context and prevent information loss. It provides core features such as five-way RRF retrieval, confirmation-based writes, Fernet-encrypted storage, session transcripts, and maintenance tools. It supports custom path configuration via Cordis, relies on Node and Python environments, and is suitable for building long-term, structured, and secure agent memory systems.
Read Moredsh-project-memory: Project-level persistent memory plugin for deepseek-harness
`dsh-project-memory` is a zero-dependency memory plugin for deepseek-harness, designed to address the issue of information not being preserved across sessions. It writes project-stable knowledge into Markdown files in the local `memory/` directory (with `MEMORY.md` as the index), facilitating manual review, Git auditing, and version control. The plugin provides four core tools: `memory_load_context` for loading memory, `memory_upsert` for writing or updating, `memory_finalize` to solidify task conclusions, and `memory_forget` to delete expired records (while protecting the index file). Features include sensitive information interception (e.g., passwords, tokens), workspace-aware root directory resolution, and r...
Read More