Baselines Before Architecture: Evaluating Coding Agents for Autonomous Penetration Testing

Recent autonomous penetration testing papers report high benchmark scores while adding multi-component security harnesses around frontier LLMs. Because these systems often change both architecture and backbone model, it is difficult to tell how much performance comes from the harness rather than...

PT-2026-59552

PraisonAI’s MCP Model Context Protocol integration allows spawning background servers via stdio using user-supplied command strings e.g., MCP"npx -y @smithery/cli ...". These commands are executed through Python’s subprocess module. By default, the implementation forwards the entire parent proces...

PT-2026-59391

Model Import Overwrites Any Model Without Ownership Check Affected Component Model import endpoint: - backend/open webui/routers/models.py lines 254-308, import models Affected Versions Current main branch commit 6fdd19bf1 and likely all versions with model import functionality. Description The...

PT-2026-59162

Stored Cross-Site Scripting XSS in the genai/ evals visualization component of Google Cloud Vertex AI SDK google-cloud-aiplatform versions from 1.98.0 up to but not including 1.131.0 allows an unauthenticated remote attacker to execute arbitrary JavaScript in a victim's Jupyter or Colab environme...

Rethinking MCP Security: A Large-Scale Study of Runtime MCP Servers and Security Scanner Reliability

The Model Context Protocol MCP has rapidly established itself as a standard interface for enabling LLM-based agents to interact with external tools and services. As MCP servers are increasingly entrusted with security-sensitive operations, understanding their real-world risks has become critical...

PT-2026-59376

Summary The /responses endpoint in the OpenAI router accepts any authenticated user and forwards requests directly to upstream LLM providers without enforcing per-model access control. While the primary chat completion endpoint generate chat completion checks model ownership, group membership, an...

PT-2026-59380

Summary An application-wide Cross-Site Request Forgery CSRF vulnerability was found Open-WebUl's image uploading functionality. An attacker can set an image URL to a malicious endpoint, allowing them to perform actions on behalf of a victim user. Any authenticated user can exploit this...

PT-2026-59680

Summary Amazon SageMaker Python SDK is an open-source library for training and deploying machine learning models on Amazon SageMaker. An issue exists where, under certain circumstances, the Triton inference handler deserializes model artifacts without performing integrity verification, allowing...

PT-2026-59730

Impact Constructing a tflite model with a paramater filter input channel of less than 1 gives a FPE. Patches We have patched the issue in GitHub commit 34f8368c535253f5c9cb3a303297743b62442aaa. The fix will be included in TensorFlow 2.12. We will also cherrypick this commit on TensorFlow 2.11.1...

PT-2026-59167

Summary Graphiti versions before 0.28.2 contained a Cypher injection vulnerability in shared search-filter construction for non-Kuzu backends. Attacker-controlled label values supplied through SearchFilters.node labels were concatenated directly into Cypher label expressions without validation. I...

PT-2026-59237

Description The LangSmith SDK's prompt pull methods pull prompt / pull prompt commit in Python, pullPrompt / pullPromptCommit in JS/TS fetch and deserialize prompt manifests from the LangSmith Hub. These manifests may contain serialized LangChain objects and model configuration that affect runtim...

PT-2026-59304

A vulnerability in the create model version handler of mlflow/server/handlers.py in mlflow/mlflow versions 3.9.0 and earlier allows an unauthenticated remote attacker to read arbitrary files from the server's filesystem. The issue arises when a CreateModelVersion request includes the tag...

PT-2026-59309

In mlflow/mlflow versions up to 3.9.0, the SearchModelVersions REST API endpoint and the mlflowSearchModelVersions GraphQL query lack proper per-model authorization checks when basic authentication is enabled. This allows any authenticated user to enumerate all model versions across all registere...

PT-2026-59398

Stored XSS to Account Takeover via Model Profile Images in Open WebUI Affected: Open WebUI tags. On the output side, users.py added a MIME allowlist check and X-Content-Type-Options: nosniff. The fix was applied to UserUpdateForm, UpdateProfileForm, and later to ChannelWebhookForm. Three models...

PT-2026-59361

Summary Several direct, index-addressed Ollama proxy routes accept a caller-supplied url idx path parameter and use it as a raw index into the admin-configured OLLAMA BASE URLS list. Access control on these routes validates only whether the user may use the requested model, never which backend th...

PT-2026-59335

Summary Null pointer dereference SIGSEGV in Upsample 6 7::adapt upsample 6 7 onnx/version converter/adapters/upsample 6 7.h:31 when convert version processes a model with an Upsample node that has zero inputs. The adapter accesses node-inputs0-sizes without checking input count. 107-byte PoC...

PT-2026-59308

MLflow 3.9.0 with basic-auth --app-name basic-auth fails to enforce authorization checks for multiple Gateway API 'list' endpoints. Specifically, the BEFORE REQUEST HANDLERS dictionary in mlflow/server/auth/ init .py does not include entries for ListGatewaySecretInfos, ListGatewayEndpoints, and...

PT-2026-59254

LiteLLM prior to 1.83.10 allows a user to modify their own user role via the /user/update endpoint. While the endpoint correctly restricts users to updating only their own account, it does not restrict which fields may be changed. A user who can reach this endpoint can set their role to proxy...

PT-2026-59679

Summary Amazon SageMaker Python SDK is an open-source library for training and deploying machine learning models on Amazon SageMaker. An issue exists where, under certain circumstances, the ModelBuilder/Serve component stores an HMAC signing key in cleartext as a container environment variable,...

MindReader: Using LLMs to Encourage Memorable and Secure Password Replacement

We report on the design and evaluation of MindReader, a tool that helps a user replace her password when she is required to do so. Left to their own devices, users tend to replace their previous passwords with predictable variations of the original ones. MindReader leverages LLMs to suggest...