DSH workflow plugins chain steps together in DeepSeek Harness — hooks that fire on events, scheduled runs, multi-stage pipelines, and repeatable task templates.
Repositories with traceable evidence of a DeepSeek Harness relationship. They have not been reviewed, install tested, or security checked — the signal below each one is the whole of what the registry currently knows.
AI SQL 权限校验拦截插件 — DeepSeek Harness 通用 AI 数据库权限管控插件。 拦截智能体（Agent）生成的越权 SQL：不静默改写、直接拒绝执行，并把结构化权限 原因返回给模型，让模型自主剔除越权字段、更换统计口径，或用自然语言向用户解释权限不足。 插件 ID：dsh-plugin-sql-security-validate 适用框架：DeepSeek Harness（Cordis） 适配版本：DSH >= 0.1.x（本仓库验证于 0.2.0-rc.1） 开源协议：MIT
Unofficial Windows file-recovery guard for unattended DeepSeek Harness workflows
DeepSeek Harness plugin: run subagents on a local llama.cpp model via the OpenAI Responses (/v1/responses) protocol — LLM adapter + delegation skill
DietrichGebert/ponytail — 懒惰 senior 模式，hook注入,迁移成dsh
Role-based model routing for DeepSeek Harness subagent delegation.
DSH (DeepSeek Harness) 定时提醒与重复任务插件 — 让 agent 按计划主动提醒你
dsh-delegate: model-aware subagent delegation for DeepSeek Harness — per-call models, depends_on dependency gating, per-child personas, a durable run roster, audit events, and conversation-flow tool cards. | 给 DeepSeek Harness 的子代理委派加上：按次选模型、依赖门控、角色人设、任务花名册。
DeepSeek Harness 常驻插件:在会话标题栏显示主代理与每个子代理实际使用的模型和思考档位 | Persistent DSH web plugin: shows the model & reasoning effort actually used by every subagent, right in the conversation header
DeepSeek Harness 多智能体协同开发插件，内置七级智能体（岐伯、隶首、风后、夔、力牧、皋陶、离朱），覆盖项目初始化、代码归类、计划编排、批量执行、代码审查与测试验证的完整开发闭环。基于 @deepseek-ai/cordis 插件体系，为 dsh 提供“风后力牧”式多智能体协作能力。
Workflow-net runtime for DeepSeek Harness: resource-aware concurrency, native loops and fan-out, static soundness checking before a plan becomes durable, and process mining over its own event log.
dsh-subagent-rules 子代理模型与思考强度规定：subagent_flash 锁定 flash 模型路由，思考强度默认 max 且可按会话自定义，分发规则自动注入所有对话。零依赖 DSH 宿主平面插件。
Deepseek harness mermaid流程图渲染插件
