AI Agent Hub
Back to plugins
🖥️

dsh-session-export

Client Updated 2026.08.17

Run the following command in DeepSeek Harness:

dsh plugin install uckkk/dsh-session-export

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

Run dsh plugin install uckkk/dsh-session-export in DeepSeek Harness (source: https://github.com/uckkk/dsh-session-export), then restart or reload the profile to activate it.

About this plugin

DSH sessions sit quietly on your local disk, and when you need to back up, migrate, or tidy them up, there is no convenient single step to do so. dsh-session-export closes that gap: export any local session to the format you need, without digging through raw log files by hand.

The plugin ships two core tools. session_list enumerates every local session by title, working directory, and creation time, with keyword filtering for quick lookup; session_export renders the chosen session as Markdown (preserving user and assistant messages, tool calls with results, and reasoning steps) or as structured JSON for downstream pipelines. It natively reads both zstd-compressed logs (.jsonl.zstd, the DSH default) and plain-text logs (.jsonl), and is a pure Node implementation with zero third-party runtime dependencies.

Whether you want a regular backup of conversation history, a polished Markdown export for team docs or a blog post, or a JSON dump to feed a knowledge base for summarization and analytics, dsh-session-export covers the need with minimal setup. Paired with dsh-session-search for cross-session full-text search and dsh-context-dashboard for token and context stats, it forms a complete local session management toolkit.

Use Cases

  • Regularly back up DSH conversation history to prevent data loss
  • Export a deep debugging session as Markdown for team docs or a blog post
  • Dump sessions to JSON for knowledge-base ingestion, summarization, or analytics

Best For

  • Developers who need to archive or migrate DSH sessions
  • Engineers who prefer Markdown for technical notes
  • Knowledge workers building personal knowledge bases with structured session data