EXPLOIT-CVE-2026-33634

CVE-2026-33634 — LiteLLM supply chain + SSRF in apibase PoC / Lab ⚠️ Deliberately vulnerable lab, for educational and authorized use only. Run it only on your machine, against the containers in this repository. Read SECURITY-NOTES.md before starting. 🚫 NEVER run this lab on a cloud VM or a shared...

CVE-2026-105148

A vulnerability was identified in SciPhi-AI R2R up to 3.6.6. This vulnerability affects unknown code of the file py/shared/abstractions/llm.py of the component Retrieval Completion API Endpoint. Such manipulation of the argument generationconfig.apibase leads to server-side request forgery. The...

PYSEC-2026-4070 LiteLLM Proxy has server-side request forgery via the `user_config` request parameter

Summary A server-side request forgery in LiteLLM Proxy lets an authenticated caller redirect the proxy's outbound request to a host of their choosing by smuggling an apibase inside the userconfig request body, bypassing the existing parameter guard. Details LiteLLM Proxy validates request bodies...

LiteLLM Proxy has server-side request forgery via the `user_config` request parameter

SummaryA server-side request forgery in LiteLLM Proxy lets an authenticated caller redirect theproxy's outbound request to a host of their choosing by smuggling an apibase inside theuserconfig request body, bypassing the existing parameter guard. DetailsLiteLLM Proxy validates request bodies with...

GHSA-HX8V-G79F-8W5F LiteLLM Proxy has server-side request forgery via the `user_config` request parameter

Summary A server-side request forgery in LiteLLM Proxy lets an authenticated caller redirect the proxy's outbound request to a host of their choosing by smuggling an apibase inside the userconfig request body, bypassing the existing parameter guard. Details LiteLLM Proxy validates request bodies...

CVE-2026-59823

LiteLLM is a proxy server AI Gateway to call LLM APIs in OpenAI or native format. Prior to 1.83.9, an authenticated LiteLLM Proxy caller with a valid virtual key can place apibase inside the userconfig request body to bypass isrequestbodysafe, which blocks top-level apibase and baseurl but...

CVE-2026-59823

LiteLLM is a proxy server (AI Gateway) for calling LLM APIs. Prior to version 1.83.9 , an authenticated caller holding a valid virtual key could place api_base inside the user_config request body, bypassing the is_request_body_safe check that blocks top-level api_base and base_url but did not ins...

CVE-2026-59823: Server-Side Request Forgery (SSRF)

LiteLLM is a proxy server AI Gateway to call LLM APIs in OpenAI or native format. Prior to 1.83.9, an authenticated LiteLLM Proxy caller with a valid virtual key can place apibase inside the userconfig request body to bypass isrequestbodysafe, which blocks top-level apibase and baseurl but...

PYSEC-2026-3865 MLflow AI Gateway permits SSRF through an unvalidated api_base

MLflow's AI Gateway accepts an authconfig.apibase value when creating a gateway secret mlflow/server/handlers.py, creategatewaysecret with no validation of scheme, host, or IP range; the value is stored verbatim. The gateway proxy endpoint mlflow/server/gatewayapi.py, rawproxy subsequently issues...

MLflow AI Gateway permits SSRF through an unvalidated api_base

MLflow's AI Gateway accepts an authconfig.apibase value when creating a gateway secret mlflow/server/handlers.py, creategatewaysecret with no validation of scheme, host, or IP range; the value is stored verbatim. The gateway proxy endpoint mlflow/server/gatewayapi.py, rawproxy subsequently issues...

CVE-2026-84377: Server-Side Request Forgery (SSRF)

LiteLLM is a proxy server AI Gateway to call LLM APIs in OpenAI or native format. Prior to versions 1.88.6 and 1.96.2, any authenticated LiteLLM proxy user could redirect an outbound provider call to a destination the user controls and cause the proxy to send its configured provider credentials t...

CVE-2026-71211

A flaw was found in MLflow's AI Gateway where an unvalidated apibase value in a gateway secret could lead to a Server-Side Request Forgery SSRF vulnerability. An authenticated user, even with read-only privileges, can create a secret with an internal network address. This allows the gateway proxy...

MLflow AI Gateway permits SSRF through an unvalidated api_base

MLflow's AI Gateway accepts an authconfig.apibase value when creating a gateway secret mlflow/server/handlers.py, creategatewaysecret with no validation of scheme, host, or IP range; the value is stored verbatim. The gateway proxy endpoint mlflow/server/gatewayapi.py, rawproxy subsequently issues...

GHSA-H7X2-H6G9-P789 MLflow AI Gateway permits SSRF through an unvalidated api_base

MLflow's AI Gateway accepts an authconfig.apibase value when creating a gateway secret mlflow/server/handlers.py, creategatewaysecret with no validation of scheme, host, or IP range; the value is stored verbatim. The gateway proxy endpoint mlflow/server/gatewayapi.py, rawproxy subsequently issues...

CVE-2026-4035: Insertion of Sensitive Information Into Sent Data

A vulnerability in mlflow/mlflow versions prior to 3.11.0 allows for the resolution of environment variables in AI Gateway secrets, which can be exploited to exfiltrate sensitive server-side environment credentials to an attacker-controlled endpoint. This issue arises because the apikey field in...

CVE-2026-34936

PraisonAI is a multi-agent teams system. Prior to version 4.5.90, passthrough and apassthrough in praisonai accept a caller-controlled apibase parameter that is concatenated with endpoint and passed directly to httpx.Client.request when the litellm primary path raises AttributeError. No URL schem...

CVE-2026-34936

PraisonAI is a multi-agent teams system. Prior to version 4.5.90, passthrough and apassthrough in praisonai accept a caller-controlled apibase parameter that is concatenated with endpoint and passed directly to httpx.Client.request when the litellm primary path raises AttributeError. No URL schem...

CVE-2026-34936

Summary (CVE-2026-34936): PraisonAI prior to 4.5.90 allowed SSRF through passthrough/apassthrough by using a caller-controlled api_base that is concatenated with an endpoint and sent to httpx.Client.request() when a fallback path triggers an AttributeError. No URL scheme validation, private IP fi...

CVE-2026-34936: Server-Side Request Forgery (SSRF)

PraisonAI is a multi-agent teams system. Prior to version 4.5.90, passthrough and apassthrough in praisonai accept a caller-controlled apibase parameter that is concatenated with endpoint and passed directly to httpx.Client.request when the litellm primary path raises AttributeError. No URL schem...

CVE-2024-12779

A Server-Side Request Forgery SSRF vulnerability exists in infiniflow/ragflow version 0.12.0. The vulnerability is present in the "POST /v1/llm/addllm" and "POST /v1/conversation/tts" endpoints. Attackers can specify an arbitrary URL as the "apibase" when adding an "OPENAITTS" model, and...