dsh-lens
Run the following command in DeepSeek Harness:
dsh plugin install hajimixiaojie/dsh-lens
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install hajimixiaojie/dsh-lens in your terminal to install this plugin; the source is available at https://github.com/hajimixiaojie/dsh-lens
About this plugin
Typing mistakes in TypeScript or JavaScript—a missing bracket, a type mismatch—often do not surface until compile time or CI. dsh-lens brings the code health check right up to the moment you edit: after every save it runs a fast syntax scan, and when the AI assistant reads or modifies a source file it quietly kicks off a full type-check and lint pipeline in the background. No manual commands required.
Every issue found is collected into a single diagnostic card displayed above the DSH chat input, while the same findings are shared with the AI assistant so it can spot errors and offer targeted fixes. The plugin never blocks your edits, never modifies files on its own, and card refreshes go through a dedicated read-only endpoint so snapshot data stays out of the conversation context and the chat remains clean.
If you are building in TypeScript or JavaScript inside DSH web interface and want code-quality checks to become a seamless background habit, or you would prefer the AI to be aware of existing issues before it starts rewriting code, dsh-lens is built for exactly that workflow.
Use Cases
- Saving a TypeScript file instantly surfaces syntax or type errors in a diagnostic card
- AI assistant receives background lint and typecheck findings before acting on code
- Eliminates manual tsc or eslint runs; saving a file triggers a full check
Best For
- Developers writing TypeScript or JavaScript inside the DSH web interface
- Teams wanting to shorten the edit-compile-debug loop without extra tooling
- Users who want the AI assistant to see code issues before rewriting
Related Plugins
A unified suite combining hot runtime injection, task-aware thinking-mode routing, and a graded session protocol with red-team gates to sustain model diligence across long-horizon inference.
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.