CVE-2024-1561 — AI Deep Analysis Summary

🚨 **Essence**: Gradio allows calling **any method** on classes via endpoints. 📉 **Consequences**: Unauthorized **Local File Read (LFR)**. 💥 **Impact**: Sensitive information leakage from the server.

🛡️ **Root Cause**: Improper access control on endpoints. 🔍 **CWE**: **CWE-29** (Improper Control of Filename for Include/Include-Adjacent). ⚠️ **Flaw**: No restriction on which class methods can be invoked remotely.

👥 **Affected**: **Gradio** (Python library). 📅 **Versions**: **4.3 to 4.12**. 🚫 **Fixed In**: Version **4.13.0** and above.

🕵️ **Hackers Can**: Read arbitrary files from the server. 🔑 **Privileges**: Local file system access. 📂 **Data**: Sensitive configs, source code, secrets, or user data.

🔓 **Threshold**: **Low**. 🌐 **Auth**: Likely **None** required for the specific endpoint. ⚙️ **Config**: Depends on Gradio deployment, but the flaw is in the core method invocation logic.

🔥 **Public Exp?**: **Yes**. 📜 **PoC**: Available on GitHub (DiabloHTB). 🧪 **Tools**: Nuclei templates exist for automated scanning. 🌍 **Exploitation**: Active proof-of-concepts are public.

🔍 **Self-Check**: Use **Nuclei** with CVE-2024-1561 template. 🛠️ **Command**: `nuclei -target <URL> -t CVE-2024-1561.yaml`. 📋 **Verify**: Check if your Gradio version is < 4.13.0.

✅ **Fixed**: **Yes**. 📦 **Patch**: Upgrade to **Gradio 4.13.0+**. 🔗 **PR**: Fixed in PR #6884. 🔄 **Action**: Update library immediately.

🚧 **No Patch?**: Implement strict **input validation**. 🚫 **Block**: Restrict method names allowed in API calls. 🛡️ **WAF**: Block requests targeting internal Gradio endpoints.…

⚡ **Urgency**: **HIGH**. 🚨 **Priority**: Critical for ML model deployments. 📢 **Action**: Patch immediately if running vulnerable versions. ⏳ **Risk**: Easy exploitation leads to data breach.