dsh-compact-activity
Run the following command in DeepSeek Harness:
dsh plugin install wallpap/dsh-compact-activity
Paste the following prompt into your AI chat to install this plugin:
Install this plugin by running dsh plugin install wallpap/dsh-compact-activity in the DeepSeek Harness terminal. Source: https://github.com/wallpap/dsh-compact-activity
About this plugin
When working with DeepSeek Harness for complex tasks, the model's thinking process and tool calls can generate numerous intermediate steps, which vertically occupy significant screen space in the Web interface, requiring frequent scrolling to view the complete conversation or final output and reducing browsing efficiency. The dsh-compact-activity plugin addresses this by intelligently grouping consecutive thinking and tool calls into a single collapsible unit. By default, these process items are condensed into one line, displaying only a summary and status indicator (such as in progress, completed, or error), with details available upon expansion; the plugin features a soft visual design with color-coded states and maintains compatibility with official components, ensuring no alteration to underlying data. This plugin is particularly beneficial for developers, data scientists, and AI researchers who frequently use DeepSeek Harness for agent workflows and need to efficiently browse and debug long sequences of tool calls in conversations, as well as for any user looking to optimize their workspace, reduce visual clutter, and navigate more cleanly through dialogues.
Screenshots
Use Cases
- Reduce scrolling distance when browsing long conversations.
- Monitor tool call status in real-time.
- Switch interface language in multilingual environments.
Best For
- Developers using agent workflows.
- Data scientists debugging tool call sequences.
- AI researchers browsing complex conversations.
Related Plugins
A method pack that makes AI coding agents plan against your real baseline, prove completion with fresh evidence, and reduce reworks and unsafe changes.
Turns the DeepSeek Harness session into a captain that builds a durable sub-agent team, splits goals into dependency-aware tasks, and coordinates work via direct messages and a live Web UI.
Gives coding agents design judgment, letting Claude Code, Cursor, and 70+ agents generate and iterate high-quality UI, presentations, and graphics right from the terminal.
Run the Pi ecosystem's plugins on DeepSeek Harness, unmodified, via a compatibility layer that implements Pi's public extension ABI on DSH's native services.