dsh-move-session
Run the following command in DeepSeek Harness:
dsh plugin install hucj09/dsh-move-session
Paste the following prompt into your AI chat to install this plugin:
Install this plugin in DeepSeek Harness by running dsh plugin install hucj09/dsh-move-session; the source code is available at https://github.com/hucj09/dsh-move-session.
About this plugin
Moving a conversation between workspaces is a recurring need in any multi-workspace dsh setup. The official fork creates a new session that inherits the current default model rather than the source session's actual model and presets, and offers no option to archive the original. dsh-move-session adds a Move Session entry to the session header and the sidebar context menu, filling that gap precisely. Migration performs an event-level full copy of an idle session: every message, tool call, title, agent preset, provider/model, and request/header event is preserved verbatim. The copy inherits the source session's last request/header model and agent preset, and lineage is recorded via parentSession in the same semantic as the official fork. After the move you choose whether to keep or archive the original, and the sidebar updates across all open tabs in real time without a page reload. If the target workspace already contains a session with the same name, the copy receives an auto-incremented suffix such as [MS1] or [MS2] to avoid collisions.
The plugin is dependency-free, modifies zero lines of dsh source, and ships as-is with no build step. It is built for developers who maintain multiple workspaces and need to relocate sessions between scratch and production areas; users who want a true move (copy plus optional archive) rather than copying and then manually cleaning up; and teams that demand full integrity of session logs, tool-call records, and model configurations down to the last event.
Core capabilities at a glance: cross-workspace event-level full copy, model and preset inheritance, keep-or-archive-the-original choice, automatic same-name disambiguation, real-time sidebar refresh across tabs, idle-session-only safety check, and target-directory pre-validation that prevents orphan copies before any write occurs.
Screenshots
Use Cases
- Relocate a finished session from scratch to production and archive the original
- Reuse a full conversation with tool calls and model configs across projects
- Auto-inherit the source session model and preset so no reconfiguration is needed
Best For
- Developers managing multi-workspace dsh environments
- Users requiring event-level preservation of session logs and model settings
- Teams that prefer true session moves over copy-and-manual-cleanup workflows
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.