CVE-2026-31233: Guardrails AI code injection in Hub installer
Attack on AI2 Disclosed vulnerabilitySupply chainInfrastructure vulnerability
Guardrails AI contains a code injection vulnerability in its Hub package installation mechanism due to unsanitized script execution from manifests. Attackers can publish malicious packages to execute arbitrary code on victim systems.
disclosed CVENo confirmed exploitation, so this record is tracked as a disclosed vulnerability and not counted as an incident. It moves into the counts if CISA or the vendor confirms exploitation. Why
Details
- type
- Attack on AIAn AI model, agent or AI-powered product was the thing exploited, abused or leaked.
- categories
- Supply chainprimaryMalicious or backdoored models, poisoned packages, unsafe deserialization and other compromised AI dependencies.
- Infrastructure vulnerabilityCVEs and advisories in AI frameworks, inference servers, MCP servers, AI developer tools and hosting stacks.
- Supply chainprimary
- owasp llm top 10
- mitre atlas
- Both rows are derived from the type and categories (OWASP 2025 edition, ATLAS release 2026.09), not classified per record. How
- severity
- 2 · Disclosed vulnerabilityCVE or vendor advisory, patched or patchable, no confirmed exploitation.
- targets
- Framework / tooling
- products
- Guardrails AI
- incident date
- 12 May 2026
- first reported
- 12 May 2026
- origin
- GitHub Advisory DatabaseVulnerability record from NVD / the GitHub Advisory Database. This product uses the NVD API but is not endorsed or certified by the NVD.
- classifier confidence
- 0.95 · published at or above 0.80; see the methodology
- tags
- CVE-2026-31233 · code-injection · rce
- record
- created 2026-10-18 07:00Z · updated 2026-10-18 07:00Z · json
Sources · 1
- 2026-05-12CVE-2026-31233: Guardrails AI contains a code injection vulnerability in its Hub package installation mechanismGitHub Advisory Database · github.com/advisories/GHSA-r6hf-g5x6-7pv9
Titles and links belong to their publishers. The summary above is written by GCD AI RADAR in its own words; data is CC BY-SA 4.0.
Related incidents
- 2026-06-10CVE-2026-47155: vLLM artifact pin decay allows loading unpinned code and weights2 Disclosedon AIInfra vulnSupply chain2 src
- 2026-05-28vLLM Hardcoded Trust Remote Code Parameter Leads to RCE2 Disclosedon AIInfra vulnSupply chain1 src
- 2026-05-22CVE-2026-5817: Docker Model Runner vllm-metal arbitrary code execution vulnerability2 Disclosedon AIInfra vulnSupply chain1 src
- 2026-05-08CVE-2026-44843: LangChain vulnerable to unsafe deserialization2 Disclosedon AIInfra vulnSupply chain2 src
- 2026-05-07Hugging Face Diffusers custom pipeline trust_remote_code bypass vulnerability2 Disclosedon AIInfra vulnSupply chain1 src
- 2026-05-07Hugging Face Diffusers trust_remote_code bypass allows remote code execution2 Disclosedon AIInfra vulnSupply chain1 src