dsh-session-link
Run the following command in DeepSeek Harness:
dsh plugin install jinhuang712/dsh-session-link
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install jinhuang712/dsh-session-link inside DeepSeek Harness to install the plugin, whose source code is hosted at https://github.com/jinhuang712/dsh-session-link
About this plugin
Sharing context across DSH sessions has meant exporting a log, ferrying the file around, unpacking zstd-compressed JSONL, and sifting through raw event streams line by line. dsh-session-link collapses that entire pipeline into a single click: tap the header button and a pure-ASCII @session- reference lands on the clipboard, ready to paste into any other session.
The reference renders as a blue chip in both the composer and sent messages. On the model side, the session_read tool resolves the reference and projects the event stream into clean user:/assistant: dialogue rows, skipping chunk, tool, and reasoning noise — a 7 MB log typically reads back as roughly 60 KB of usable conversation. Inputs accept exact IDs, dsh-session: URIs, or @title lookups, with optional maxChars budget and head or tail truncation controls.
Built for developers who routinely move conclusions between DSH sessions: one session does the thinking, the other builds on the result, and the context flows like a normal message instead of a manual file-transfer ritual.
Screenshots
Use Cases
- Session A reaches a conclusion; Session B references it directly without manual log export and transfer
- Share background context across parallel DSH sessions and skip the export-unzip-read ritual
- Model reads a full conversation via session_read, compressing a 7 MB raw log into ~60 KB of readable dialogue
Best For
- Developers who frequently work across multiple DSH sessions in parallel
- AI engineers who need to reuse conversation context between sessions
- Technical users who want to cut down manual log-processing overhead
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.