@bigbigtooth/dsh-devops：在 DSH Web 里做 SSH 运维与 Git 部署闭环
dsh-devops（@bigbigtooth/dsh-devops）是 DeepSeek Harness 的 Web 插件，在侧边栏 Remote Operations 中管理 SSH 主机，监控硬件、进程与日志，并以只读 AI 巡检分析异常；部署侧提供首次 AI 编排、脚本固化与 git pull --ff-only 的一键更新闭环。凭据 AES-256-GCM 加密，巡检由工具权限强制只读。安装需先设置 DSH_DEVOPS_KEY_FILE，使用 dsh plugin --profile web add 后重启 DSH 主机。MIT 许可证，维护者 bigbigtooth。
Read More项目容器化适配：用 Agent Skill 把仓库落成可部署的 Docker 方案
介绍 SkillHub 社区 Skill「项目容器化适配」（@indiv-zhudev/projcet-containerd）。它用 SKILL.md 约定六阶段流程：分析仓库与配置、容器内打包制品、生成运行镜像与 compose、外置配置模板，并输出 deploy 目录下的脚本与文档。文章说明适用场景、语言覆盖与限制，给出 skillhub CLI 安装命令及 Cursor 等客户端的 skills 目录示例，以及 compose 打包与启动的典型命令。资料来自 SkillHub 开放 API 与官方安装文档。
Read More@weibaohui/dsh-git-server: DeepSeek Harness Native Git Service Plugin
DeepSeek Harness (DSH)’s `dsh-git-server` plugin integrates a complete self-service Git service into the native interface, fulfilling the “everything is a plugin” design philosophy. The plugin supports HTTP transport and end-to-end management of repositories, issues, pull requests, and Wikis, while providing engineering capabilities such as branch protection, tiered permissions, and account integration. Users can create repositories, browse content, and collaborate directly from the native management pages. It also supports building closed-loop workflows through features such as Webhooks and automatic issue closure. The plugin depends on `user-management`, supports one-click start/stop and data backup, and enables developers to use Git services efficiently within the DSH ecosystem without additional configuration.
Read MoreIntroducing the DeepSeek Harness (DSH) Plugin: dsh-fde-tools
`weibaohui/dsh-fde-tools` is an aggregator plugin for DeepSeek Harness (DSH), designed to simplify the management of multiple plugins. It bundles 19 commonly used plugins into a single installation package, offering a comprehensive suite of capabilities including a Git server, knowledge base, team collaboration, Agent management, file sharing, and a visual dashboard. After installation, these features are centrally managed via the "FDE Toolbox" panel. By utilizing a bootstrapper to handle mechanisms such as dependency conflicts, build interception, and snapshot restoration, the plugin ensures a stable installation process and significantly lowers the barrier for users to extend DSH workflows.
Read MoreDeepSeek Harness Autopilot Plugin Overview
DeepSeek Harness Autopilot (DSH Autopilot) is an operations management plugin for DeepSeek Harness (DSH), designed to address the challenges of continuous and stable agent execution. The plugin automatically executes approved tickets and review-ready pull requests. Through inbound control, queue scheduling, worktree execution, lifecycle management, and state validation, it ensures workflows operate under control before human review and CI decisions. It supports pause, resume, and cancel operations and provides Web-based observability. Users must first install DSH, add the plugin via the command line, and configure MCP servers (such as GitHub and Jira) in the configuration file to bind trackers and code repositories. The plugin is currently in the Alpha stage and does not replace DSH itself or
Read Morejackson-chen97/dsh-devops: GitLab and Kubernetes DevOps Console
The DeepSeek Harness DevOps (DSH) plugin is designed to address efficiency issues in interactions between AI agents and CI/CD systems as well as Kubernetes clusters. It provides a visual Web console that supports GitLab project management (MRs and pipelines) and Kubernetes cluster operations (Pod status, logs, and rollbacks), with built-in background monitoring and alerting. It also supports multi-server configuration and internationalization. Installation via local checkout is recommended. Once configured, the plugin can be used interactively through conversation to directly invoke AI tools for automated operations.
Read MoreDeepSeek Harness Plugin: ddtcorex/dsh-maestro-govard
`ddtcorex/dsh-maestro-govard` is a lightweight bridge plugin for DeepSeek Harness (DSH), designed to integrate Govard’s tool capabilities into DSH sessions without embedding Go code. It registers shell-invocation tools to support environment start/stop, database operations, and read-only deployment checks (generating Plan documents), while ensuring path resolution consistent with the Maestro suite. Installation requires the `govard` binary to be available. The plugin runs with the current permissions and is suitable for scenarios requiring fine-grained control over infrastructure and deployment workflows.
Read MoreDeepSeek Harness Plugin: DSH Get
The DeepSeek Harness (DSH) Get plugin is designed to address fragmented plugin management workflows by integrating search, installation, and auditing capabilities into the DSH interface. This community plugin supports offline searches by name, tags, and other criteria, and provides features such as installation controls, shell injection protection, and the output of six audit evidence items. Users must install it via the command line and restart DSH. Configuration files allow adjustment of behavior parameters, and users should carefully review the source code of third-party plugins when using it to ensure security.
Read Moredsh-kubectl: kubectl Integration Plugin for DeepSeek Harness
dsh-kubectl is an open-source plugin that encapsulates kubectl capabilities into the DeepSeek Harness (DSH) framework, providing agents with structured, confirmation-based interactions for Kubernetes operations. The plugin supports querying resources, viewing logs, port forwarding, and executing write operations with user confirmation, addressing the issue that conventional output is often not agent-friendly and can be parsed unreliably. Installation requires using the patch command and configuring the kubectl path, while write operations rely on explicit confirmation to prevent accidental execution.
Read Moredsh-runbook-ops: DeepSeek Harness Workflow Orchestration Plugin
dsh-runbook-ops is a plugin for DeepSeek Harness designed to address the challenges of validating, reusing, and auditing process definitions. It provides Standard Operating Procedures (SOPs) as structured YAML and does not replace the Harness execution system; instead, it generates actionable guidance. The plugin supports listing, reading, and validating runbooks, as well as controlling execution. After installation, the runbook directory is specified via a configuration file, and final execution relies on Harness’s existing tools and approval system, ensuring process security and auditability.
Read More# Sentry Skill: Integrating Production Error Troubleshooting into AI Agent Workflows
Sentry is an Agent Skill maintained by OpenAI in the curated directory of the openai/skills repository, which enables read-only observability queries via the official Sentry CLI: listing unresolved issues, viewing event details, and calling issue explain/plan to generate root cause analysis and repair suggestions. This article introduces its positioning, core commands, installation methods for both the CLI and the Skill (Codex's $skill-installer and Cursor's .cursor/skills), as well as precautions including the default 24-hour production environment query scope, JSON output format and security desensitization, which is suitable for DevOps/SRE to integrate online troubleshooting into AI workflows.
Read More### 正式译法（适配技术文档场景）： Setting up CI: Let AI Agents Help You Build GitHub Actions CI/CD Pipelines
# setup-ci from awesome-cursor-skills: A beginner-friendly DevOps skill for GitHub Actions This skill detects Node.js, Python, Go and other projects via the SKILL.md specification, then generates a `ci.yml` workflow that includes linting, testing, type checking and optional deployment. It emphasizes the use of `npm ci`, dependency caching and secret security best practices. This article covers installation (deploying to `.cursor/skills/`), typical usage and precautions, and is ideal for quickly setting up CI/CD for new projects.
Read MoreGitHub Stacked PR Public Beta: Split large changes into PR chains that can be independently reviewed and merged with one click
On July 30, 2026, GitHub announced that Stacked Pull Requests had entered public beta, supporting the splitting of large changes into an ordered PR chain, independent layer-by-layer reviews, and one-click merging of the entire stack. Based on official changelogs and documentation, this article introduces the branch dependency model, review and CI mechanisms, the getting-started steps for the gh-stack CLI, as well as the collaboration methods with Merge Queue and Copilot Agent and current limitations.
Read MoreCopilot Code Review Integrates Agent Skills and MCP: Team Guidelines Finally Can Be Included in Reviews
On July 29, 2026, GitHub announced the general availability of Copilot code review support for Agent Skills and MCP servers, which is now available to Pro, Business and Enterprise users. Teams can inject internal coding and review specifications via the SKILL.md file in the .github/skills directory; MCP can pull external contexts such as Jira and document systems in read-only mode, and attribution tags for skills and MCP are added to review comments. This article sorts out the core capabilities of the general release, the configuration examples of SKILL.md, the read-only security boundary of MCP and implementation suggestions.
Read More