FuncWei/dsh-wechat-mp-studio
FuncWei★ 0Python最后同步: 2026-08-16
微信公众号内容生产工作台:防低创作度结构轮换写作法 + 视觉基线 + gpt-image 配图管线 + OCR 验收 + 小绿书草稿接口实测 (DeepSeek Harness plugin)
README 摘要
dsh-wechat-mp-studio WeChat Official Account content studio for DeepSeek Harness 中文 English The first open-source plugin to package the anti-“low creativity” playbook for WeChat Official Account (公众号) content production — distilled from a real account that got flagged by the platform, went through remediation, and recovered. Not theory. The problem Recurring-content accounts (morning greetings, goodnight wishes, solar-term cards, homilies, blessing cards…) rarely fail because they “write badly”. They fail because the platform flags them as low creativity and turns off public recommendation. The platform scores a combination of signals — category × format × frequency × fixed phrases: - The same greeting text on every image, the same title prefix, the same hashtags, day after day; - Multiple near-duplicate posts per day cannibalizing each other; - Templated structure: five posts with the same skeleton = low-creativity signal. This plugin provides an executable anti-homogenization production pipeline : structure-signature rotation, the “daily delta” test, a visual baseline for blessing images, OCR acceptance for on-image text, interaction rules, and draft-box sync — including measured…
在 GitHub 查看完整 README →分类
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). | 全网最强 DeepSeek Harness 外挂视觉插件,为 DeepSeek、GLM 等纯文本模型外挂视觉能力,粘贴图片即得结构化 JSON 证据(OCR、版面、语义)。
★ 2,208
Anionex/agent-vision-toolkit为纯文本模型"看图“设计更好的视觉工具箱和技能,支持多图理解,图片问答,前端UI还原、GUI 自动化等,并可选无缝接入多个主流agent,直接识别粘贴图片| A vision toolkit and skill designed for text-only llms — image Q&A, long-screenshot OCR, frontend UI restoration, and GUI automation, with optional seamless integration for Codex, Claude Code, Pi, Oh My Pi, and OpenCode
★ 933
Anionex/dsh-vision-toolkit让纯文本模型更好地做视觉任务的DeepSeek Harness插件:带意图的图片问答、长截图 OCR、UI 还原等|DeepSeek Harness-native integration for agent-vision-toolkit: image Q&A, long-screenshot OCR, UI restoration, grounding, pixel diff, Artifacts, and Web UI.
★ 488