Ai-gateway-payload-processing: ai-gateway-payload-processing: tenant/credential selection from client-controlled url path with no in-process authorization
- Published
- Last modified
- Status
- published
- Known exploitation
- Not listed in CISA KEV
- Public exploit
- None on Exploit-DB
Summary
The ai-gateway-payload-processing component, which routes requests to large language models for multiple tenants, is affected. An authenticated remote attacker could use another tenant's model identifier to have that tenant's API keys injected into their request, allowing them to make unauthorized calls that consume the other tenant's LLM resources and quota.
Summarized by anthropic/claude-haiku-5.5. Verify details against the original sources.
Read the original description
A flaw was found in ai-gateway-payload-processing. An authenticated remote attacker can exploit this vulnerability by submitting a request specifying another tenant's model identifier in the request body or header. Because the service does not verify authorization before retrieving credentials from the shared store, it injects the target tenant's application programming interface (API) keys into the request. This flaw allows an attacker to perform unauthorized requests and consume another tenant's large language model (LLM) resources and quotas.
Affected versions
- rhoai/odh-ai-gateway-payload-processing-e2e-rhel9
Version range not reported
- rhoai/odh-ai-gateway-payload-processing-rhel9
Version range not reported