In DeepSeek Harness (DSH) development practice, we modularize all functionality. When calculating discrete probability distributions in an agent or workflow, directly integrating large math libraries is often not flexible enough. Below is an introduction to a plugin specifically designed for calculating the probability mass function (PMF) of the Poisson distribution.
What It Is¶
dsh-poisson-pmf is a DSH plugin maintained by user uckkk. It provides Poisson distribution probability mass function calculation capabilities and belongs to the model reasoning category. The plugin implements this mathematical logic in pure Node.js, making it suitable for scenarios in agents or workflows that require handling discrete probability distributions.
Core Features¶
- Pure Node.js implementation
- No network dependencies
Installation and Enablement¶
To install the plugin, simply run the following command:
dsh plugin add github:uckkk/dsh-poisson-pmf
Notes¶
- Permission-based execution: The plugin runs with the permissions of the current DSH process.
- Dependency requirements: Before use, ensure that
@deepseek-ai/cordis(>= 4.0.1) and@deepseek-ai/dsh-tools(>= 0.1.0-rc.6) are installed. - Source code review: It is recommended to check the source code and license before installation. This plugin is licensed under the MIT License.
Summary¶
This plugin addresses the need to quickly introduce pure mathematical computation logic in the DSH ecosystem without relying on bulky external libraries. For more details, see: GitHub.