RAGFlow: OpenAI-compatible agent completion bypasses private canvas authorization

Summary

RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine. From 0.25.2 until 0.27.3, POST /api/v1/agents/chat/completions invokes completion_openai and returns through the openai-compatible branch before UserCanvasService.accessible() checks whether the caller may access the requested agent. An authenticated user who knows another user's private agent UUID can execute that Only me workflow and read its output across the user or tenant boundary; configured tools may also cause side effects, but stronger impacts were not demonstrated. This issue is fixed in version 0.27.3.

Remediation

1. Inventory every infiniflow ragflow deployment, including staging.
2. No fixed version appears in the public sources yet. Check the references below and the vendor directly.

References

• Referencehttps://github.com/infiniflow/ragflow/security/advisories/GHSA-c6q8-23rp-32cv
• Referencehttps://github.com/infiniflow/ragflow/pull/19817
• Referencehttps://github.com/infiniflow/ragflow/commit/20180790e1a8c55553a2f3e75718f3d65178ae6d
• Referencehttps://github.com/infiniflow/ragflow/releases/tag/v0.27.3