AIKIDO-2026-571058
@openai/agents-core is vulnerable to Improper Authorization
59
Medium Risk
This Affects:
@openai/agents-coreTL;DR
The hostedMcpTool helper and tool-search loading in the core package build the hosted MCP require_approval policy without validating its shape. Unsupported policy strings, unknown object keys, malformed toolNames filters, and tools listed in both always and never are accepted and passed through unchanged. A malformed or contradictory approval policy can silently weaken or disable the human-in-the-loop approval gate so hosted MCP tool calls run without the intended approval. The fix adds a shared validator that normalizes the policy and rejects invalid or overlapping configurations.
Who does this affect?
You are affected if you are using a version that falls within the vulnerable range and configure hosted MCP tools with require_approval approval policies.
Background info
@openai/agents-core is vulnerable to Improper Authorization in versions 0.0.7 - 0.10.0.
How to fix this
Upgrade the @openai/agents-core and/or the @openai/agents library to the patch version.
Are You Affected?
Connect your repositories to instantly see whether vulnerable or malicious packages exist in your codebase.
Free. No credit card required.
SOC 2Compliant
ISO 27001Compliant