dsh-pin-recall
Run the following command in DeepSeek Harness:
dsh plugin install kerwin2046/dsh-pin-recall
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install kerwin2046/dsh-pin-recall inside DeepSeek Harness to install; the source is at https://github.com/kerwin2046/dsh-pin-recall .
About this plugin
In long multi-turn sessions, a critical tool output or assistant conclusion can easily get buried under subsequent turns. dsh-pin-recall lets you pin any reply, tool result, or free-form text directly in the session and recall it back into the model context with a single command, eliminating manual copy and paste.
The interaction works on two layers. In the Web UI, every finished assistant message shows a star toggle at the bottom; a corner-icon button then packages all pins and wakes the agent in one click. On the command side, /pin, /pins, /recall, and /unpin cover per-message, per-type (tool or assistant), and free-text granularity, with a quiet-queue option that enqueues without triggering a wake.
Built for developers running deep reasoning chains, debugging tool calls, or repeatedly referencing early conclusions inside DeepSeek Harness, so the evidence trail stays one recall away.
Use Cases
- Reusing early tool outputs across a long reasoning chain
- Quickly recalling key conclusions while debugging multi-turn agents
- Anchoring assistant replies as context for subsequent turns
Best For
- Developers running long reasoning chains in DeepSeek Harness
- Agent debuggers who repeatedly reference early conclusions
- Engineers mixing Web UI and CLI 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.