dsh-code-quote
Run the following command in DeepSeek Harness:
dsh plugin install hanrr92/dsh-code-quote
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install hanrr92/dsh-code-quote in the DeepSeek Harness terminal; the source repository is https://github.com/hanrr92/dsh-code-quote. Restart the profile after installation to activate the plugin.
About this plugin
Pasting code blocks with a path:line header into the DSH input box floods the draft with raw multi-line text—hard to scan and ambiguous for the model. dsh-code-quote solves this by collapsing such snippets into a single compact reference chip, keeping the input clean while still delivering the full code context to the model.
Once pasted, the code folds into a backspace-able reference chip (file icon + filename:line) that renders identically to built-in @path chips in the chat stream. On send, an agent/pre-step hook appends the full code snapshot—captured at the moment of folding—as a standalone context message, visible as an expandable context injection row in every subsequent turn. Snapshots persist to a local JSON file, surviving profile restarts.
Ideal for developers who reference local code snippets frequently in DSH conversations and want a tidy input experience without losing context fidelity. It does not modify code content—only its presentation and delivery—and should be enabled in place of (not alongside) dynamic reference plugins to avoid duplicate injection.
Use Cases
- Pasting local code snippets into DSH for model analysis and debugging
- Keeping the input box and chat chips clean when referencing lines from multiple files
- Ensuring the model retains the exact code content at fold time rather than the latest file state
Best For
- Developers who frequently reference local code snippets in DSH conversations
- Daily users who want a clean input box while ensuring full code context reaches the model
- Advanced workflow users who need code references to persist across profile restarts
Related Plugins
Memory layer for coding agents that indexes local session history and auto-recalls relevant context before edits, commands, and failures, with no manual search needed.
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.