AI Agent Hub
Back to plugins
dsh-ssh-files-sidebar preview

dsh-ssh-files-sidebar

Workflow Updated 2026.09.04

Run the following command in DeepSeek Harness:

dsh plugin install qigelunbiya/dsh-ssh-files-sidebar

Paste the following prompt into your AI chat to install this plugin:

Run dsh plugin install qigelunbiya/dsh-ssh-files-sidebar in your DeepSeek Harness terminal to install; full source is at https://github.com/qigelunbiya/dsh-ssh-files-sidebar

About this plugin

When doing remote development inside DeepSeek Harness, you are often juggling between a file sidebar, a terminal, a local workspace, deployment notes, and release scripts. Taking over an unfamiliar project makes things harder because there is no clear path from reconnaissance to a stable run. dsh-ssh-files-sidebar brings all of these steps into a single Conversation and a single SSH security boundary, so file browsing, code editing, terminal work, and even reading a server screenshot for vision or OCR happen in one workflow instead of copying images back to your local machine.

What truly sets the plugin apart is its deployment-maturity path. For projects with no deployment documentation, the Agent first performs read-only checks on the local tech stack and the remote environment, then executes a step-by-step Bootstrap Plan until the service is genuinely stable, and only then distills the actual successful path into a human-readable DEPLOYMENT.md. After you confirm which stages you want to hand off to a machine, an automation script is generated. Once the Runbook and the script have both been validated in real use, the Agent upgrades from issuing commands to executing releases, running health checks, diagnosing issues from logs, and, with your explicit approval, rolling back. Every step stays inside a strict security boundary: one session binds to one SSH target, and high-risk operations still go through Harness native approval.

It is built for developers who frequently operate Linux servers inside Harness, who write code locally but start services and debug on the remote box, and for small-to-medium engineering teams that want to turn tribal deployment knowledge into documentation, automation, and ultimately an Agent-driven release loop.

Screenshots

Use Cases

  • Connect to an SSH server in Harness, browse and edit remote files, and check logs in a terminal—all in one conversation
  • Take over an unfamiliar project and let the Agent research the local stack and remote environment, bootstrap a working deployment, and produce a DEPLOYMENT.md
  • After the Runbook is validated, generate a one-click script and let the Agent execute releases, run health checks, diagnose from logs, and roll back with approval

Best For

  • Engineers who frequently operate Linux servers in DeepSeek Harness for remote development and debugging
  • Engineering teams that want to turn tribal deployment knowledge into documentation and automation scripts
  • Technical leads aiming to evolve from manual deployments to an Agent-driven release loop