Install and configure the native PaddleOCR plugin for DeepSeek Harness (DSH) from the Settings → PaddleOCR GUI. Use for OCR and image-to-text from screenshots, scans, and PDFs; Chinese/CJK text; PDF-to-Markdown; structured document parsing with tables, formulas, layout, and reading order; or DSH endpoint, credential…
A design workflow for producing self-contained HTML or SVG files that open directly in a browser. It covers visual work such as posters, infographics, landing pages, charts, banners, cards, and interface mockups.
Instructions for writing and debugging dsh-hooks-plugin scripts, which run during tool or session events such as before a tool call or after it finishes. These scripts return JSON decisions such as allowing, asking about, or denying an action.
Use when DeepSeek Harness needs bounded, reproducible collection of posts, creator pages, first-level comments, or nested comments from Xiaohongshu, Douyin, Kuaishou, Bilibili, Weibo, Tieba, or Zhihu through a separately installed MediaCrawler checkout. Prefer ordinary web search for quick facts or a few…
Use when a user asks to find, compare, or evaluate an existing DeepSeek Harness plugin, Skill, MCP server, agent extension, profile, or similar ecosystem capability.
Dispatch and monitor an explicitly scoped, simple or medium-complexity local subtask through models available in DeepSeek Harness, including read-only review and user-authorized workspace implementation. Use when the user says “调用 Harness”, asks Harness to analyze or review, or delegates a fix, edit, refactor, test…
A set of rules for choosing the right kind of DeepSeek Harness extension before building it. It explains the differences between bundles, local patches, agent presets, skills, and temporary runtime plugins.
A meta-skill for finding and safely evaluating an outside agent skill when the skills already installed do not cover a task. It treats outside instructions as untrusted material until they have been reviewed.
Use when creating new skills, editing existing skills, or verifying skills work before deployment - skill TDD: pressure scenarios, unaware subagents, fail-then-fix-then-pass loops.
Diagnose live Kubernetes failures (CrashLoopBackOff, OOMKilled, NodeNotReady, ImagePullBackOff, Network/CNI failures, HPA starvation, DNS failures, pending pods) with an evidence-first procedure: capture context, collect signals (events/status/logs), rank falsifiable hypotheses, verify one variable at a time, fix with…
A read-only security audit for DeepSeek Harness extensions, settings, dependencies, and runtime exposure. It looks for risks such as exposed secrets, unsafe file mounts, data leaving the system, persistence, and malware indicators.
Use when auditing DSH or cross-harness skills, presets, hooks, tools, commands, MCP configuration, or other agent assets for discovery, routing, integrity, and safe ownership. Do NOT use for code-style review, raw session analysis, or repository outcome scoring.
A structured system for legal research involving laws, court decisions, academic sources, arguments, and evidence. It separates facts, legal claims, scholarly views, and the author's own analysis.
A guided project-initialization skill that interviews the user about a project and fills its AGENTS.md and documentation placeholders. AGENTS.md is a handbook that tells coding agents how to work in a repository.
Use when building, extending, or fixing tools/plugins for DeepSeek Harness (dsh) — writing a dsh plugin entry module, registering a tool with defineTool, authoring parameters/output JSON Schemas, the dsh bundle manifest, or cordis.patch.yml, or when a plugin fails to load, a tool is missing from the agent, or tool…
CoAligne workflow skill. Use when a user wants an Agent to use shared CoAligne project context: inspect synced files, read Agent sessions, understand teammate activity, send files or sessions to teammates, download cloud files, resolve CoAligne file/session links, or troubleshoot CoAligne Desktop sync.
★not rated 2 21d agoA66 tokens
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: