Introduction¶

In the DeepSeek Harness (DSH) plugin ecosystem, handling text deduplication, fuzzy matching, or similarity judgment is a common requirement. Directly calling external API services increases latency and generates additional costs. The dsh-string-similarity plugin implements localized text similarity calculation using pure Node.js, addressing comparison needs in offline or private environments.

Plugin Overview¶

This plugin is maintained by uckkk, belongs to the model inference category, and is released under the MIT open-source license. It provides string similarity calculation capabilities, enabling developers to complete text comparison tasks locally without network dependencies.

Core Features¶

• String similarity calculation
• Supported algorithms: Levenshtein, Jaccard
• Runtime: pure Node.js implementation, no network or external services required
• Invocation: directly call the tools registered by the plugin in a session

Installation¶

Run the following command to install the plugin:

dsh plugin add github:uckkk/dsh-string-similarity

Usage¶

After installation, directly call the tools registered by this plugin in a session.

Applicable Scenarios and Notes¶

• Suitable for scenarios that require local text similarity comparison.
• Note: installing the plugin runs third-party code on your local machine. Please review the source code yourself.

Conclusion¶

This plugin provides DSH users with a lightweight, offline solution for string similarity calculation. For more details, refer to the plugin directory or source repository.