dsh-clamp
Run the following command in DeepSeek Harness:
dsh plugin install uckkk/dsh-clamp
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install uckkk/dsh-clamp in DeepSeek Harness to install this plugin; the source code is available at https://github.com/uckkk/dsh-clamp
About this plugin
In model-inference pipelines, output values occasionally drift outside the range you actually need, and unchecked drift can break downstream logic. dsh-clamp addresses exactly that: it provides a set of numeric clamping tools that keep model outputs within the bounds you define.
Under the hood it is a pure Node implementation with zero network dependencies. The footprint is small, startup is instant, and once installed the tools are registered directly in the session with no extra services or complex configuration.
If you are building inference pipelines with hard numeric boundaries on output, or simply need a quick local way to verify a set of clamp rules, this plugin slots into your workflow with minimal overhead. Its MIT license means you are free to use it in any project.
Use Cases
- Model output drifts outside the expected range and needs quick convergence
- Validate a set of clamp rules locally without spinning up extra services
- Embed lightweight numeric boundary constraints into an inference pipeline
Best For
- Developers building inference pipelines with hard numeric bounds on output
- Engineers who need zero-network-dependency local clamping tools
- Users who want to call lightweight post-processing tools directly in a session
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.