dsh-difference
Run the following command in DeepSeek Harness:
dsh plugin install uckkk/dsh-difference
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install uckkk/dsh-difference in your DeepSeek Harness terminal to install this plugin (source: https://github.com/uckkk/dsh-difference).
About this plugin
In model-inference pipelines, taking the difference of two candidate sets is a frequent yet tedious step: dropping items already seen in a previous round, comparing two generations, isolating new elements. dsh-difference packages this single operation into a ready-to-call tool, implemented in pure Node.js with zero network calls and zero external dependencies, so you get the result on the spot without spinning up a throwaway script.
The capability is intentionally narrow: supply two sets and receive the elements present in the first but absent from the second. There is no configuration layer, no dependency tree, no abstraction to learn. Once registered as a session tool, you pass in the data and get the output back, all locally with no HTTP request involved.
It suits developers who need a quick set-comparison inside an inference session and would rather not leave the conversation, open a terminal, and write a one-off Node script just to compute a difference. Call the tool, get your subset, move on to the next reasoning step.
Use Cases
- Dropping candidates already produced in the previous inference round
- Diffing two model outputs to pinpoint added or missing elements
- Computing a subset inside a session without leaving the conversation to write a throwaway script
Best For
- Developers who need a fast set comparison inside an inference loop
- Engineers who want zero-dependency, zero-network local computation
- Developers who would rather not spin up a one-off Node script just to compute a difference
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.