dsh-pluck
Run the following command in DeepSeek Harness:
dsh plugin install uckkk/dsh-pluck
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install uckkk/dsh-pluck in the DeepSeek Harness terminal to install, source at https://github.com/uckkk/dsh-pluck
About this plugin
During model-inference conversations it is common to need a single field value buried inside a long model output. Repeatedly prompting the model to output only that field wastes tokens and risks format drift. dsh-pluck pulls this step out of prompt engineering and turns it into a deterministic local tool call.
Built purely on the Node runtime with zero network dependencies, the plugin registers as an in-session tool once installed. You describe the field path you need, and it resolves the value locally without any network round-trip or extra inference tokens.
Well suited for teams and developers whose inference pipelines frequently involve extracting intermediate values from chain calls, parsing form-structured outputs, or any routine where regenerating the full output just to grab one field feels wasteful.
Use Cases
- Extracting a single field value from a long model output
- Parsing specific fields from intermediate results in chained inference
- Quick field lookup in form-structured or structured outputs
Best For
- Developers who frequently process model inference outputs
- Teams looking to reduce inference token consumption
- Engineers building local inference workflows
Related Plugins
A unified suite combining hot runtime injection, task-aware thinking-mode routing, and a graded session protocol with red-team gates to sustain model diligence across long-horizon inference.
ModLens is a vision plugin for DeepSeek Harness that gives text-only models sight by reading images pasted directly into chat, with zero-config setup and multiple vision engines.
On-demand vision for text-only DeepSeek Harness agents: built-in free keyless vision chain and 14 vision tools, routing image turns as tool calls to vision models with pixel fidelity, no Python needed, one-command install.
Give text-only models in DeepSeek Harness eyes, enabling image Q&A, long-screenshot OCR, UI restoration, and GUI visual tasks.