AI Agent Hub
Back to plugins
🧠

dsh-session-delete

Memory Updated 2026.08.15

Run the following command in DeepSeek Harness:

dsh plugin install TANGBINGBINGING/dsh-session-delete

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

Run dsh plugin install TANGBINGBINGING/dsh-session-delete in your terminal; the full source is available at https://github.com/TANGBINGBINGING/dsh-session-delete .

About this plugin

DeepSeek Harness lets you archive or delete a conversation from the sidebar, but once that happens, finding a specific session, restoring in bulk, or purging permanently quickly becomes tedious. dsh-session-delete turns the archive area and the trash into two commands — /t-archive and /t-restore — so the full lifecycle (list, restore, delete, clear) stays at your fingertips.

The interaction model is built around stable list numbering: between two refreshes a number always maps to the same conversation, and an operated slot shows a gap instead of renumbering. Combine that with keyword-first targeting and batch number selection, and working through dozens or hundreds of archived sessions becomes far less error-prone. The plugin is also multi-workspace aware — the archive listing spans every workspace folder, restores return each conversation to its original directory, and the trash is global with a single clear-all.

It is aimed at heavy Harness users who run several workspaces in parallel, frequently spawn fork or sub-agent sessions, and want file-manager-style control over conversation history. The plugin ships as a self-registering bundle, so no manual loader configuration is needed; restart dsh and type the command.

Use Cases

  • Archive area has grown large; filter by keyword and batch-restore or purge in one pass
  • Accidentally deleted a conversation; retrieve it from the global trash and restore it to its original workspace folder
  • Working across multiple workspaces; browse every archive folder and clean up orphaned fork or sub-agent sessions

Best For

  • Heavy Harness users running multiple workspaces in parallel with large conversation volumes that need frequent archiving and cleanup
  • Researchers or teams whose workflows spawn many fork and sub-agent sessions that must be tracked and cleaned
  • Users who want file-manager-style control over conversation lifecycle via stable numbering, keyword targeting, and batch operations