Ragflow API Client

Universal client for Ragflow API enabling dataset management, document upload, and running chat queries against self-hosted RAG knowledge bases.

This skill appears to be what it claims: a Node-based Ragflow API client. The primary risk is operational: the tool will upload files and send them to whatever RAGFLOW_URL you provide using the RAGFLOW_API_KEY. Only install/use it if you trust the Ragflow instance and you provide a least-privilege API key. Also note the registry metadata lists no required env vars while SKILL.md and the code require RAGFLOW_URL and RAGFLOW_API_KEY — verify that metadata mismatch before automating credential injection. Finally, avoid running the CLI in contexts where untrusted users could supply arbitrary file paths to upload sensitive data.

Type: OpenClaw Skill Name: ragflow Version: 1.0.2 The skill is designed to interact with a RAGflow API, including uploading documents. Its `uploadDocument` function in `lib/api.js` and the `upload` command in `scripts/ragflow.js` allow reading and uploading arbitrary local files specified by a `filePath` argument (e.g., `--file /etc/passwd`). While this functionality is inherent to a document upload feature, the lack of input sanitization or restrictions on the `filePath` creates a significant vulnerability. If an AI agent is susceptible to prompt injection, it could be instructed to upload sensitive local files (e.g., `~/.ssh/id_rsa`, `/etc/passwd`) to the configured `RAGFLOW_URL`, leading to local file exfiltration. This is a critical vulnerability that allows attacks, rather than code explicitly designed with malicious intent.