When building applications based on DeepSeek Harness (DSH) or developing agents, you often need to perform structured analysis on conversation history or text content, such as counting the frequency of specific terms. This usually requires developers to write additional scripting logic. The DSH plugin mechanism allows this capability to be directly extended.
The following is an introduction to a plugin focused on local text analysis. This plugin is maintained by uckkk, and its core function is word frequency statistics.
Plugin Features¶
- Pure Node implementation
- No network or external services required
Installation and Configuration¶
First run the installation command:
dsh plugin add github:uckkk/dsh-word-frequency
After installation, in the package.json file of the current profile, add "dsh-word-frequency" to the dsh.profile.bundles field.
Dependency Requirements¶
This plugin depends on the following environment:
@deepseek-ai/cordis: ^4.0.1@deepseek-ai/dsh-tools: >=0.1.0-rc.6
Usage¶
After installation and configuration, you can call the tool registered by this plugin in the session.
Notes¶
Installing the plugin means running third-party code on your local machine. Before using it, please review the source code and license (MIT) yourself.