dsh-duration
Run the following command in DeepSeek Harness:
dsh plugin install uckkk/dsh-duration
Paste the following prompt into your AI chat to install this plugin:
Visit https://github.com/uckkk/dsh-duration to review the source code, then run dsh plugin install uckkk/dsh-duration in your terminal to add the duration formatting plugin to your DeepSeek Harness setup.
About this plugin
In model-inference pipelines, timing data often shows up as raw millisecond or second values that are hard to parse at a glance. dsh-duration tackles this small but annoying problem: it normalizes duration values into readable, human-friendly formats so that inference logs, session outputs, and performance reports tell a clear story.
The tool is a pure Node.js implementation that runs entirely on the local machine. It makes no network calls and depends on no external services, which means there is no added latency, no flakiness from connectivity issues, and no extra infrastructure to maintain.
If you are building inference workflows on DeepSeek Harness and need to surface token-generation time, batch-processing duration, or any other timing metric in a way that end users or log readers can quickly understand, dsh-duration is a lightweight, zero-dependency companion. Install it, register it in your profile bundle, and invoke the tool directly in your session with no further configuration required.
Use Cases
- Convert raw millisecond values in inference logs into human-readable durations
- Present token-generation and batch-processing timings with a consistent, readable format in session output
- Render timestamp deltas in performance reports as concise, easy-to-read durations
Best For
- Developers who need to display inference timing data to end users
- Engineers building DeepSeek Harness workflows with a focus on output readability
- Ops and SRE teams who want time fields in logs and reports to be immediately legible
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.