DEDUPED Hot AI repos, turned into use cases Get the Monday 5

How to pick calendar slots and save them to the task with DeepSeek Harness Genui

Choose writing blocks or calendar slots in an interface and keep the choice for later steps.

Credentials. Any calendar action needs task-scoped access from Harness, and you should not place calendar credentials or secrets in generated-app state.

Who it is forA person scheduling focused work who needs to compare several connected time options before committing.

The problemChoosing among many connected time slots in plain text is slow and easy to get wrong.

The outcomeA saved set of chosen slots in the task, with any later calendar action kept separate and gated by permission.

What you need first

  • Node.js ^22.19.0 or ^24.0.0
  • A supported DeepSeek Harness Web profile
  • Access to the dsh CLI to add the plugin

Steps

  1. dsh plugin --profile web add dsh-plugin-genui
  2. dsh --profile web
  3. Start a new Web session and describe the slots or writing blocks you want to compare.
  4. Select the blocks you want in the interface and save them to the task.
  5. Ask the Agent to continue from the saved selection, and approve any calendar action separately when Harness asks.

Prompt for your agent

Use the dsh-plugin-genui plugin in DeepSeek Harness to build a React + TypeScript app where I select from <list of time slots> and save the picks to the task. Output the saved selection and confirm that a later calendar action is a separate step that needs permission.

Caveat. The README does not describe how calendar connections are set up, so declare any connected tool before expecting the Agent to call it.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for pengyue-polaron/deepseek-harness-genui and "Pick calendar slots and save them to the task"

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

Share: X LinkedIn Hacker News Email Copy link

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer