dsh-outline: Search, Read, and Safely Write to the Outline Knowledge Base in DSH
`dsh-outline` is a Web plugin for DeepSeek Harness (DSH), designed to address the challenge of interacting directly with an Outline knowledge base during conversations. It provides core capabilities such as searching, reading full articles, controlled writing, and local saving. API tokens can be configured through GUI cards, and it supports multi-user usage and security mechanisms such as HTTPS, retries, and caching. Installation is simple, making it suitable for developers who need to manage a knowledge base within conversations.
Read Moredsh-helloai-works: DeepSeek Harness Card Knowledge Base
The DeepSeek Harness memory plugin supports managing text and image content using tags and cards. Card content can be organized into blocks, with support for renaming, tag merging, and content search. Users can input `@` in a conversation to quickly reference individual cards or tag groups. The plugin stores images using SHA-1 and features automatic saving, optimistic locking, and a snapshot mechanism, with data persisted in a local folder. Installation requires the official command, and data remains available after uninstallation.
Read Moredsh-kb: A Team Knowledge Base Plugin for DeepSeek Harness
The DeepSeek Harness plugin dsh-kb aims to address fragmented team knowledge by providing long-term knowledge accumulation and management across sessions and team members. Based on the Karpathy Wiki model, it organizes knowledge into three layers: raw (immutable materials), wiki (written knowledge), and schema (conventions). Users can access the full knowledge base from the sidebar, with support for switching between multiple libraries, custom processing rules, full-text search, version history, and automatic distillation. Unlike dsh-file-share, which only manages session files, dsh-kb focuses on maintaining a long-term, fixed knowledge base. Installation requires DeepSeek Harness version 0.1.7-rc.2 or higher.
Read Moredsh-learn-wiki: A Learn-by-Doing Knowledge Base Plugin
`dsh-learn-wiki` is a CRAG-style corrective retrieval plugin in the DeepSeek Harness ecosystem for filling knowledge gaps after static retrieval failures. It uses a three-tier architecture: L1 (Markdown knowledge base), L2 (Hindsight), and L3 (Web). When retrieval misses, the plugin automatically performs rate-limited background web supplementation based on struggle-detection signals, stores the new knowledge in a staging area, and then uses explicit tools to review and commit it to the knowledge store. Implemented in pure JavaScript with no build step required, the plugin supports explicit retrieval, conversation history extraction, and knowledge provenance.
Read Moredsh-siyuan: 17 Tools for Siyuan Note Integration
The DeepSeek Harness plugin `dsh-siyuan` integrates with SiYuan Notes, supporting note retrieval, reading, writing, and management through 17 `siyuan_*` tools. Users can install it via commands and configure the API token, default notebook, and permissions on a dedicated settings page. Read-only and daily note features are enabled by default, while write and risky operations must be manually enabled. This plugin provides a low-barrier local knowledge base integration for DSH and requires DSH 0.1.7 or later.
Read MoreDeepSeek Harness Plugin: Search and Read the Outline Knowledge Base in Conversations
`dsh-outline-auto` is a web plugin for DeepSeek Harness (DSH) designed to connect the Outline knowledge base with AI conversations. The plugin provides 13 tools supporting keyword search, full-text reading, local document saving, and document management (create, delete, and update operations require user approval). It includes security and fault-tolerance mechanisms such as read caching, retries for HTTP 429 errors, enforced HTTPS, and user credential isolation, meeting enterprise knowledge base integration requirements.
Read MoreDeepSeek Harness Plugin: dsh-tiddlywiki
The DeepSeek Harness (DSH) plugin dsh-tiddlywiki integrates TiddlyWiki 5 into the DSH ecosystem, enabling humans and agents to share a plain-text knowledge base. Data is stored using Git, supporting version control, audit trails, and portability at any time. The plugin features a “read-before-write” safety mechanism, optimistic concurrency control, and restart protection to ensure write safety. Users can read and write notes through the interface, and agents can directly invoke tools to operate on the same data, without vector retrieval or background scraping, enabling AI to participate in knowledge management while keeping users in full control.
Read Moredsh-knowledge: DeepSeek Harness Knowledge Base Plugin with Local and Remote Dual Backends
# Introduction Using DSH (DeepSeek Harness) for daily development, an unavoidable issue arises: project standards, architectural decisions, and lessons learned occur within specific sessions and vanish once the session ends. The typical solution is to manually copy important content into system prompts, but this leads to increasingly bloated contexts, making it impossible to control retrieval scopes and write permissions. The dsh-knowledge introduced below turns this into a standalone component: without modifying the DSH Agent Loop, the same plugin can utilize local SQLite or connect to a remote centralized knowledge base, providing on-demand retrieval, controlled backfilling, and a Web management console. This article proceeds in the order of features, installation, and typical usage.
Read MoreDeepSeek Harness Plugin: dsh-llm-wiki
dsh-llm-wiki is a zero-runtime-dependency plugin based on DeepSeek Harness (DSH), built with a pure ESM architecture and requiring Node.js 18 or above. By invoking the LLM-Wiki engine CLI, it encapsulates seven core tools for agents: retrieval, page reading, statistics, validation, fixing, error log viewing, and data ingestion. This plugin aims to implement the philosophy of "everything is a plugin," supporting direct knowledge base queries within conversations, ingesting source texts into the knowledge base, and continuously maintaining the knowledge base structure. Among these, query-type tools require no additional LLM configuration, while data ingestion and the fix functionality with Finalize require configuring an OpenAI-compatible LLM API. The installation process includes installing the Python engine via pip, adding the plugin via the dsh command, and configuring it in the pr
Read More