Flyvhidbwo/dsh-vision-proxy
dsh-vision-proxy is a tool for DeepSeek Harness. DeepSeek Harness plugin: DeepSeek Pro brain + automatic image recognition. Images attached in the GUI are processed by default with the official deepseek-v4-flash-vision-exp native vision model, converted to text, and passed to DeepSeek for answers (even text-only V4-Pro can see images); supports any OpenAI-compatible VLM such as Bailian/Zhipu/OpenRouter; auto-detects local Ollama without a key; one-question confirmation during install
REVIEW SCORECARD
✓ No risk detected · Automated scan results are for reference only, not an official endorsement
AI DEEP REVIEW
Solid, well-documented vision bridge that lets text-only DeepSeek models handle GUI image attachments via configurable V
README cites a CI workflow badge for Node 22/24, 14 passing tests, SHA-256 caching, error classification, and graceful sharp fallback, indicating real engineering rigor.
Keys are read from config or env vars (VISION_API_KEY/DASHSCOPE_API_KEY) rather than hardcoded, and a startup PRIVACY NOTICE plus local-Ollama option are documented, though the postinstall prompt and optional sharp native build add minor supply-chain surface.
It fills a concrete gap the README identifies: GUI image attachments are rejected for text-only DeepSeek models, and this plugin transcribes them so Pro/Flash can still answer.
Last updated 2026-08-26, not archived, and the README explicitly tracks dsh 0.1.1-rc.2 adapter compatibility, showing active upkeep.
README covers install command, pnpm allowBuilds workaround, registry mirror tip, provider/model table with pricing, key precedence, and a real demo with screenshots.
Generated by AI after reading the project README — a decision aid, not an endorsement. Neutral scores are given when information is limited.
Loading GitHub README...
The GitHub README is temporarily unavailable. Please try again later.
Open repositoryINSTALL REFERENCE
dsh plugin --profile web add dsh-vision-proxyThis command passed the install-plan safety checks and can be executed by the DSH host. Validation is not a security audit or official endorsement.
PROJECT TOPICS
VALIDATION LADDER
FAQ
DeepSeek Harness plugin: DeepSeek Pro brain + automatic image recognition. Images attached in the GUI are processed by default with the official deepseek-v4-flash-vision-exp native vision model, converted to text, and pa
dsh-vision-proxy has 15 stars and 0 forks on GitHub, last updated 2026-08-26.
dsh-vision-proxy is distributed under the MIT license.
dsh-vision-proxy 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
The system checks GitHub Topics against the site taxonomy and keyword rules. Current matches:image-understanding、multimodal、ocr、vision。
liustack
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).
tong-io
TongFlow — Multimodal GenAI Studio
Anionex
[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.
DDDMUC
Free web search provider for DeepSeek Harness - DuckDuckGo backend, no API key needed
anysearch-team
AnySearch web search provider and advanced search tools for DeepSeek Harness (DSH)
omdsh-dev
Connect DSH to your database for conversational data analysis and actionable business insights.