dsh-vision-toolkit is a tool for DeepSeek Harness. [dsh] A more powerful visual toolkit for text-only models: one-line install and use, paste images for direct recognition, multi-image Q&A, screenshot-to-frontend UI restoration, and more|DeepSeek Harness-native integration for agent-vision-toolkit: image Q&A, long-screenshot OCR, UI restoration, grounding, pixel diff, Artifacts, and Web UI.
A well-documented, actively maintained vision plugin that gives text-only DSH agents real image understanding capabiliti
Code quality8
README shows a structured two-layer architecture (tools/Skill plus native DSH integration) with npm publishing and a cordis.patch.yml config, but no test or CI details are visible.
Security6
No install command beyond the npm add is shown and no permission or secret-handling details are given, so risk cannot be assessed from the provided information.
Utility9
It directly addresses a concrete pain point for text-only models by adding image Q&A, long-screenshot OCR, UI restoration, grounding, and pixel diff in one toolkit.
Maintenance9
Last updated 2026-08-25, not archived, 830 stars, and a recent-updates section indicate an actively maintained project.
Docs8
README includes highlights, a three-step quick start, toolbox, configuration/limits, and troubleshooting sections plus English and Chinese versions, though the install command field itself is marked unknown.
Generated by AI after reading the project README, as a decision aid; neutral scores are given when information is thin. Not an official endorsement.
[dsh] A more powerful visual toolkit for text-only models: one-line install and use, paste images for direct recognition, multi-image Q&A, screenshot-to-frontend UI restoration, and more|DeepSeek Harness-native integrati
How popular is dsh-vision-toolkit?
dsh-vision-toolkit has 830 stars and 37 forks on GitHub, last updated 2026-08-25.
What license does dsh-vision-toolkit use?
dsh-vision-toolkit is distributed under the MIT license.
Is dsh-vision-toolkit compatible with DeepSeek Harness?
dsh-vision-toolkit is listed in the DSH Universe directory as a tool for DeepSeek Harness. This plugin is listed in the DSH Universe directory and covered by its validation pipeline.
CLASSIFICATION EVIDENCE
/10类依据
项目类型技能
功能/10类文件与数据
规则置信度High
System reads GitHub Topics first, then compares against the in-site category dictionary and word-root rules. Current match:agent-skills、agent-vision-toolkit、computer-vision、ocr、screenshot-testing、ui-restoration、vision-language-model、vision-tools。
The first vision plugin for DeepSeek Harness, and the vision bridge for every text-only coding agent. Paste an image, get structured JSON evidence (OCR, layout, semantics). | The strongest vision add-on plugin for DeepSeek Harness, adding vision capability to text-only models like DeepSeek and GLM, paste an image and get structured JSON evidence (OCR, layout, semantics).
Local-first A-share research workbench for DeepSeek Harness: market dashboards, watchlists, valuation, four investor agents, versioned reports, and continuous post-report chat.