dsh-task-worktree
Run the following command in DeepSeek Harness:
dsh plugin install Letter2025/dsh-task-worktree
Paste the following prompt into your AI chat to install this plugin:
To install this plugin in DeepSeek Harness, run dsh plugin install Letter2025/dsh-task-worktree in your terminal; the source code URL is https://github.com/Letter2025/dsh-task-worktree.
About this plugin
In code development using DeepSeek Harness, managing complex branches and context pollution is often a major pain point. The dsh-task-worktree plugin solves this by introducing a "task-scoped" Git Worktree mechanism. It ensures every task runs on an isolated branch while keeping the main workspace clean, eliminating the chaos of mixed contexts.
Drawing inspiration from tools like Qoder and Codex, the plugin allows you to create and manage workspaces via simple commands or a UI selector. The model handles environment creation, but critical operations like committing or deleting branches remain under explicit human control. This design balances the convenience of AI automation with the security and control of traditional Git workflows.
Whether you are working on large codebases or experimenting with new features, this plugin significantly boosts efficiency. Your task states persist across sessions, and the branch badge on the session header provides clear visibility. Whether bringing changes back to the main branch or committing directly, you maintain full control, making it an ideal companion for deep AI-assisted coding.
Use Cases
- Parallelly handling multiple independent dev tasks in the same repo.
- Conducting experimental code changes without contaminating the main branch.
- Cleaning up temporary code changes when a session ends.
Best For
- Developers managing complex projects with DeepSeek Harness.
- Experimental coders needing isolated test environments.
- Team collaborators who want to keep the codebase clean.
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.