dsh-custom-subagents
Custom reusable Subagent Definitions with a Settings page and one delegate_agent tool for DeepSeek Harness
116 results
Custom reusable Subagent Definitions with a Settings page and one delegate_agent tool for DeepSeek Harness
DeepSeek Harness agent preset: minimal-mode with web search and AGENTS.md loading (极简模式(网络)).
DSH 插件配装引擎:任务 → 双检索(规则+LLM) → 组合评分(协同/冲突/成本/信任) → 配装建议。区别于目录/搜索:按任务自动配整套插件,含冲突检测与安装命令导出。
DSH 多方群聊协作插件(圆桌):多 Agent 研讨房间,@指定/全体研讨/接力链,轮次控制与一键总结
Browse and install pinned Agent Skills from Awesome DeepSeek Skills in DSH
Mi Home (米家) control for DeepSeek Harness agents — list homes/devices, read props, and control devices behind a human approval gate.
A better approach to editing files in DSH.
MQTT protocol driver and agent worker gateway for DeepSeek Harness
Windows Port Inspector Bundle for DeepSeek Harness
Interactive ACP (Agent Client Protocol) v1 plugin for the DeepSeek Harness (dsh): run dsh agents in Zed's Agent Panel as a dsh bundle plugin.
DeepSeek Harness plugin: join ACN so this agent can discover, message, and collaborate with other agents. Defaults to the China region.
DSH rules plugin: activate rule prompts / markdown documents by glob-matching the files an agent reads or edits (Claude Code rules.md style).
SLI/SLO/错误预算:定义可衡量的可靠性目标、错误预算告警、SRE 实践。受 wshobson/agents(38k★ MIT)启发。
事故手册:严重度分级/检测→分类→缓解→恢复→沟通流程/升级路径/值班交接。受 wshobson/agents(38k★ MIT)启发。
量化回测与风险指标:规避前视/幸存者偏差、考虑交易成本、walk-forward 分析、风险指标计算。受 wshobson/agents(38k★ MIT)启发。
事故复盘写作:无指责文化、根因分析、时间线、行动项,驱动组织学习。受 wshobson/agents(38k★ MIT)启发。
多 agent 团队协作:并行调试、并行功能开发、任务协调、沟通协议、团队构成。受 wshobson/agents(38k★ MIT)启发。
架构模式:Clean Architecture、Hexagonal、DDD 战术模式、依赖规则。受 wshobson/agents(38k★ MIT)启发。
微服务架构模式:服务边界、通信、数据管理、韧性模式、事件驱动。受 wshobson/agents(38k★ MIT)启发。
DSH web plugin: Cursor-like Settings pages (AGENTS.md Rules, Skills, MCP servers, import) plus a pinned 'latest input' chat bar with a backward walk
Cordis-native, evidence-driven full-stack engineering discipline for DeepSeek Harness agents
让 DeepSeek Harness 打开工作区时自动载入 Claude Code 配置:Skills/Rules/Hooks/Agents 四样资产,固定双位置(工作区自身 .claude + ~/.claude)运行时直读、零写入。Claude Code 兼容层。
Stop runaway agent loops in DeepSeek Harness: denies repeated identical tool calls and enforces a per-agent call cap, outside the model where instructions cannot reach.
数据叙事:用可视化/上下文/说服结构把数据变成推动决策的故事。受 wshobson/agents(38k★ MIT)启发。