PathClaw

通过华银康集团 PathClaw 服务对病理切片进行泛癌 AI 辅助诊断，适用于病理科医生或研究人员在收到切片图像后的初步筛查场景。当用户提供 .svs 格式的病理切片文件时，自动完成：认证登录、提交诊断任务、轮询推理结果、返回诊断结论及热力图链接。触发条件：用户请求分析病理切片、进行泛癌预测、肿瘤筛查、或提到...

Before installing or using this skill, consider the following: - Privacy: The skill will read local .svs files and upload them to https://pathclaw.pathologyunion.com. These files may contain patient identifiers; confirm you have legal/organizational permission to transmit PHI and verify the service's data handling, retention, and privacy policy. - Authentication ambiguity: The SKILL.md requires obtaining an auth token via POST /api/user/login but doesn't explain how credentials are supplied. Ask the publisher how credentials are provided and where secrets are stored — do not assume anonymous access. - Trust and provenance: The skill has no source, homepage, or publisher documentation. Prefer installing only skills backed by a known organization or official SDK and request documentation for the API endpoints used. - Safety controls: If you proceed, require explicit user consent before any upload, strip or redact non-essential metadata from slides when possible, and test with non-sensitive example files first. - Additional checks that would increase confidence: public documentation or an official SDK for PathClaw, clarification on authentication flow (where to supply credentials), a statement about PHI handling and retention, and a publisher identity or homepage.

Type: OpenClaw Skill Name: pathclaw Version: 1.0.4 The PathClaw skill is a legitimate integration for Huayinkang Group's pathology AI service. It automates the process of authenticating, uploading .svs pathology slides to 'pathclaw.pathologyunion.com', and retrieving diagnostic results. The instructions include robust file validation rules (extension, size, and readability checks) and security guidelines for handling tokens and network errors, with no evidence of malicious intent or unauthorized data exfiltration.