Remote code execution via prompt injection in Flowise Airtable Agent (CVE-2026-41265)

The `run` method of the `Airtable_Agents` class in Flowise evaluates an LLM-generated Python script without proper sandboxing. An unauthenticated remote attacker who can send prompts to a chatflow using the Airtable Agent node may use prompt injection to cause the LLM to return a malicious Python script, resulting in arbitrary code execution on the Flowise server in the context of the server process user. Affected versions: - <= 3.0.13 We recommend updating to: 3.1.0. CVSS 9.2 · our assessment: act_now Reference: https://github.com/FlowiseAI/Flowise/security/advisories/GHSA-v38x-c887-992f Details: https://vuln.today/cve/CVE-2026-41265