dsh-plugin-ollama-monitor
Run the following command in DeepSeek Harness:
dsh plugin install LouisCanBe/dsh-plugin-ollama-monitor
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install LouisCanBe/dsh-plugin-ollama-monitor in your terminal to install the plugin; the source repository is at https://github.com/LouisCanBe/dsh-plugin-ollama-monitor .
About this plugin
Running Ollama locally usually means hopping between terminal commands for status checks, speed tests, model pulls, and code evaluations, then flipping to a browser tab to read logs. dsh-plugin-ollama-monitor folds all of that into your DSH workflow: the agent side handles the logic, the web side gives you a visual panel, so you stop context-switching between shell and browser.
On the agent side, the plugin exposes a set of DSH tools: query Ollama service status, benchmark inference speed with a configurable prompt and token budget, pull models on demand, and run coding evaluations. On the web side, an interactive panel embedded in the DSH settings page lets you adjust host, default model, test prompt, numPredict, and timeout at a glance and see changes take effect immediately, without hand-editing YAML.
Built for developers who self-host Ollama and already drive their daily workflow through DSH. If you want one place to monitor, stress-test, pull models, and evaluate code quality instead of a pile of scattered CLI commands, this plugin turns those fragments into a single, panel-driven routine.
Use Cases
- Ask your agent to check Ollama service status and list available models mid-conversation
- Benchmark local model inference speed with a custom prompt and a configurable token budget
- Fine-tune test parameters from the interactive settings panel and see results instantly
Best For
- Developers who self-host Ollama and want unified monitoring and benchmarking inside their DSH workflow
- AI engineers who frequently run local model inference stress tests and code quality evaluations
- Developers who prefer a panel-driven experience over switching between terminal and browser
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.