openai is vulnerable to Uncontrolled Resource Consumption
53
Medium Risk
The SDK verifies incoming webhook signatures by splitting the client-supplied signature header on spaces and running a separate HMAC-SHA256 verification over the full signed payload for every candidate it finds. The number of candidates is never capped, so a request whose signature header packs in many values forces one payload-sized HMAC computation per value. Because the header and body reaching a webhook endpoint are attacker-controlled and require no authentication, this amplifies a single request into a large amount of cryptographic work and can exhaust CPU. The fix caps direct verifications and switches to a single constant-work HMAC comparison once the header exceeds that limit.
You are affected if you are using a version that falls within the vulnerable range and your application verifies OpenAI webhook signatures on a publicly reachable endpoint.
openai is vulnerable to Uncontrolled Resource Consumption in versions 5.8.0 - 7.5.0.
Upgrade the openai library to the patch version.
Connect your repositories to instantly see whether vulnerable or malicious packages exist in your codebase.
Free. No credit card required.

I consent to receiving marketing communications based on Aikido’s Privacy Policy.
SOC 2Compliant
ISO 27001Compliant
ISO 42001Compliant© All Intel data is openly available and commercially licensed.