dsh-when
Run the following command in DeepSeek Harness:
dsh plugin install ZhijiangTang/dsh-when
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install ZhijiangTang/dsh-when in DeepSeek Harness to install this plugin. Source code is available at https://github.com/ZhijiangTang/dsh-when
About this plugin
In Chinese-language agent workflows, users naturally say things like "tomorrow 9am" or "2 hours later", while downstream schedulers and reminder services demand a precise ISO 8601 timestamp. Manual conversion is tedious and error-prone across time zones and daylight-saving edges. dsh-when fills that gap: it registers a lightweight when tool that turns a Chinese relative-time phrase into a UTC ISO string, a local ISO string with offset, a relative-seconds delta, and the matched pattern name—all in pure ESM with zero dependencies and no build step.
The parser covers five phrase families: day offsets (today, tomorrow, day-after, optionally with a clock time), durations (N minutes, hours, days, or weeks from now), weekday expressions (next week X, auto-rolling forward when this week has passed), bare clock times (HH:MM, rolling to next day if already past), and raw ISO 8601 input. Every input follows a strict fail-fast principle: phrases outside the supported table return ok:false with a complete list of accepted forms, so a vague phrase can never silently produce a wrong timestamp and propagate into a scheduler or log.
It pairs naturally with dsh-cron-parse (periodic cron expressions) and dsh-scheduler (one-shot triggers): cron handles recurring schedules while when handles one-off moments, and the ISO output flows straight into the same scheduling pipeline. If you are building Chinese-facing agents for reminders, scheduling, or timestamped logging, dsh-when is a drop-in, zero-dependency choice.
Use Cases
- Convert colloquial Chinese time phrases to ISO timestamps directly
- Provide one-shot trigger times for scheduler plugins
- Eliminate manual time formatting in reminder and calendar workflows
Best For
- Developers building Chinese-language agent applications
- Scheduling systems that need natural-language time parsing
- Plugin users who prefer zero-dependency lightweight solutions
Related Plugins
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.
An armor-breaking plugin for DeepSeek: rewrites instruction priority, expands refusal-blocked phrases, persists across sessions, and shows a green active indicator.