AI Agent Hub
Back to plugins
🤖

dsh-writer

Model Inference Updated 2026.08.22

Run the following command in DeepSeek Harness:

dsh plugin install mario03690/dsh-writer

Paste the following prompt into your AI chat to install this plugin:

Run dsh plugin install mario03690/dsh-writer inside DeepSeek Harness to install this plugin; the source repository is at https://github.com/mario03690/dsh-writer and all 15 structured writing tools become available the moment installation completes.

About this plugin

Every time you ask a model to write a document, you are re-specifying the genre from scratch: section order, length targets, tone conventions. Ask for a changelog fifty times and you repeat the same constraints fifty times, with uneven results and wasted tokens.

dsh-writer bakes the format rules of 15 high-frequency writing tasks into ready-to-call tools: blog posts, press releases, product descriptions, FAQs, email templates, READMEs, changelogs, meeting minutes, weekly reports, cover letters, academic polishing, name ideation, SEO packs, e-commerce listings, and job descriptions. The model handles content and expression; the tool enforces structure and convention. Each call returns a polished Markdown or HTML file with correct section order, tone, and length — ready to ship without another round of formatting instructions.

It is built for content teams, indie developers, product managers, and anyone who produces formatted documents on a regular cadence. The plugin runs no local tool code; it simply issues one outbound HTTPS request through the MCP client that ships with dsh. Failed calls are never charged, documents are processed in memory and not retained, and the entire configuration is a single short YAML row you can review in under a minute.

Use Cases

  • Turn raw commit messages into a properly grouped Keep-a-Changelog entry with one call
  • Convert a meeting transcript into structured minutes listing decisions, action items, and owners
  • Generate SEO title tags, meta descriptions, and FAQ schema suggestions batch-wise from product copy

Best For

  • Content and product teams shipping polished documents on a tight cadence
  • Indie developers who want the model to focus on expression while format rules are handled elsewhere
  • Operations and marketing staff who need deliverables that skip the formatting-and-review round