dsh-ui-tools
Run the following command in DeepSeek Harness:
dsh plugin install liukj98/dsh-ui-tools
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install liukj98/dsh-ui-tools in your terminal to install; the source is available at https://github.com/liukj98/dsh-ui-tools.
About this plugin
Chaining multiple tool calls in DeepSeek Harness means arguments and return values scatter across the conversation stream, making it hard to trace what happened where. dsh-ui-tools fixes this by registering a dedicated tools tab in the conversation header that lists every tool invocation of the current session in dispatch order, so you can see at a glance who was called, with what arguments, what came back, and whether the call succeeded, failed, or is still running.
Under the hood it is a purely browser-side plugin (dsh.client) that reads nodes and runningCalls from the ConversationSnapshot. It never initiates model requests or mutates chat state. Each entry renders the tool name, formatted arguments, the full result text, and a status indicator (ok / error / running), which makes it straightforward to pinpoint a failing step in a longer tool chain.
If you routinely orchestrate multi-step tool pipelines in DSH and need to debug argument passing or audit the call chain, this lightweight view saves you from grepping logs or scrolling through the conversation repeatedly. There is no backend dependency; the tab appears as soon as dsh web restarts after installation, making it a natural fit for researchers and developers who value observability in their inference workflows.
Use Cases
- Debug multi-step tool chains by pinpointing the failing invocation
- Replay a conversation to verify tool arguments were passed correctly
- Monitor tool call status (ok, error, running) without scrolling through logs
Best For
- Developers orchestrating multi-tool pipelines in DSH
- AI researchers auditing tool pipeline execution results
- DSH users who prefer lightweight UI plugins over log scripts
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.