dsh-self-restart
Run the following command in DeepSeek Harness:
dsh plugin install NecromanAlbert/dsh-self-restart
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install NecromanAlbert/dsh-self-restart in your DeepSeek Harness terminal to install the plugin; the source code is available at https://github.com/NecromanAlbert/dsh-self-restart
About this plugin
After a DeepSeek Harness Desktop restart, any running session loses its context and the user must manually reopen the tab and re-explain the task. dsh-self-restart closes that gap: a session writes a self-contained mission handoff file before the restart, then requests the official Desktop restart.
Once the Desktop is back, the plugin reads the handoff (valid for 30 minutes), resumes the target session via agents.resume if it is not already live, injects the mission as a plugin notice into the follow-up turn, and clears the file so it is not replayed on a later boot. Only one handoff exists at a time—a second call overwrites the first—and if the host does not expose desktopActions.requestRestart, the tool still writes the handoff and asks the user to restart manually.
Best suited for long-running agent tasks that need a Desktop reboot to take effect: call one tool, describe the mission, and let the plugin handle the rest without manually reopening the browser or retyping instructions.
Use Cases
- Restart Desktop mid-task and the session auto-resumes to continue its mission
- Reboot after updating plugins or system configs without manually reopening the browser tab
- Restart across multiple sessions and resume a specific one by specifying sessionId
Best For
- Harness users running long agent tasks that require a Desktop reboot mid-run
- Heavy multi-session users who must manually reopen tabs after every restart
- Workflow automation users who want zero manual steps after a Desktop reboot
Related Plugins
Traceable, searchable cross-session memory for AI agents that turns conversation knowledge into a typed knowledge graph and recalls relevant subgraphs instead of replaying full history, natively integrated with DeepSeek Harness.
Gives DSH AI cross-session long-term memory, to-do and skill management, plus multi-session orchestration, external AI delegation, and an infinite canvas that grows with you.
dsh-mnemon is a three-tier, pluggable, Agent-driven memory system for DeepSeek Harness, combining Runtime memory, Project Documents, and replaceable Memory Spaces with nine long-term providers.
An opinionated, zero-infrastructure file-based memory protocol: human-readable markdown files plus a bounded always-loaded index, curation discipline, and no database or embeddings.