CVE-2026-107295: Cross-Site Request Forgery (CSRF)

Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 1.34.0 until 1.107.4 and 2.28.0, the Agent.toweb and clai web development chat endpoint has missing request content-type validation. A website visited by a developer can submit a...

Compromise Is Not Consequence: Evaluating Task-Scoped Authorization in LLM Agents with Paired Replay

A tool-using model can follow a malicious instruction even when its credentials are valid. We study whether task-scoped authorization contains the resulting tool execution. Our paired-replay testbed samples a model request once and submits the same action, resource, and arguments to broad bearer,...

CVE-2026-102242 Path Traversal via Symlink Following in allowedLocalRoots in MCP Toolbox for Databases

Improper link resolution CWE-59 / CWE-22 in the allowedLocalRoots path validation in Google MCP Toolbox for Databases versions 1.2.0 through 1.9.0 allows a remote authenticated attacker with tool execution permissions to bypass directory boundary restrictions via symbolic links. Because path...

CVE-2026-100544

openclaw's @openclaw/voice-call package before 2026.8.1 launches the configured agent for classic inbound voice calls without propagating the caller's identity or non-owner status. As a result, owner-only tool filtering can fail open and expose the agent's normal tool authority to a remote caller...

PT-2026-99209

Name of the Vulnerable Software and Affected Versions OpenClaw versions prior to 2026.8.1 Description A sandbox policy bypass exists in the MCP loopback component. This issue allows sandboxed coding-agent sessions to invoke tools that are explicitly restricted by the sandbox.tools.deny policy. An...

CVE-2026-89032

BerriAI LiteLLM versions prior to 1.101.0-rc.1 are vulnerable to a tenant isolation bypass within the semantic cache layer . The flaw is caused by a metadata key mismatch between the _get_semantic_cache_tenant_scope() and _get_metadata_variable_name() functions. An authenticated remote attacker w...

CVE-2026-91933

Flowise before 3.1.4 fails to enforce workspace-level authorization checks in its openai-realtime endpoints . An authenticated user can supply an unscoped chatflowid to access ChatFlows belonging to other workspaces, issuing GET and POST requests to retrieve tool definitions and execute tools fro...

CVE-2026-91933 Flowise before 3.1.4 Authorization Bypass via openai-realtime

Flowise before 3.1.4 fails to enforce workspace-level authorization checks in openai-realtime endpoints, allowing authenticated users to access tools from ChatFlows in other workspaces by supplying an unscoped chatflowid. Attackers can invoke GET and POST requests to retrieve tool definitions and...

CVE-2026-57137 PraisonAI AgentLoop onToolCall approval runs after tool execution

PraisonAI is a multi-agent teams system. From 1.4.0 until 1.7.2, createAgentLoop in src/praisonai-ts/src/ai/agent-loop.ts passes executable tools to generateText before invoking the onToolCall approval callback. Because the wrapped AI SDK executes tool handlers during generation, a callback that...

PraisonAI has an origin validation bypass in MCP HTTP Stream transport that allows browser-mediated unauthenticated tool execution on local MCP server

SummaryPraisonAI's MCP HTTP Stream transport uses an unsafe prefix match when validating the Origin header. The default localhost allowlist includes origins such as http://localhost, and the validation accepts any origin that starts with an allowed value.As a result, an attacker-controlled origin...

PYSEC-2026-3896 PraisonAI has an origin validation bypass in MCP HTTP Stream transport that allows browser-mediated unauthenticated tool execution on local MCP server

Summary PraisonAI's MCP HTTP Stream transport uses an unsafe prefix match when validating the Origin header. The default localhost allowlist includes origins such as http://localhost, and the validation accepts any origin that starts with an allowed value. As a result, an attacker-controlled orig...

CVE-2026-79750

MCPHub is a unified hub for centrally managing and dynamically orchestrating multiple MCP servers/APIs into separate endpoints with flexible routing strategies. Prior to version 1.0.30, MCPHub scopes non-admin users to servers they own list views and config edits enforce ownership, but the...

CVE-2026-79750 MCPHub authenticated horizontal IDOR: any non-admin user executes tools on other users' MCP servers (cross-tenant file read + SSRF)

MCPHub is a unified hub for centrally managing and dynamically orchestrating multiple MCP servers/APIs into separate endpoints with flexible routing strategies. Prior to version 1.0.30, MCPHub scopes non-admin users to servers they own list views and config edits enforce ownership, but the...

CVE-2026-79750 MCPHub authenticated horizontal IDOR: any non-admin user executes tools on other users' MCP servers (cross-tenant file read + SSRF)

MCPHub is a unified hub for centrally managing and dynamically orchestrating multiple MCP servers/APIs into separate endpoints with flexible routing strategies. Prior to version 1.0.30, MCPHub scopes non-admin users to servers they own list views and config edits enforce ownership, but the...

CVE-2026-79750

MCPHub is a unified hub for centrally managing and dynamically orchestrating multiple MCP servers/APIs. Prior to version 1.0.30 , while list views and config edits correctly enforce per-user ownership, the tool-execution API does not. Any authenticated non-admin user can invoke tools on MCP serve...

CVE-2026-79750 MCPHub authenticated horizontal IDOR: any non-admin user executes tools on other users' MCP servers (cross-tenant file read + SSRF)

MCPHub is a unified hub for centrally managing and dynamically orchestrating multiple MCP servers/APIs into separate endpoints with flexible routing strategies. Prior to version 1.0.30, MCPHub scopes non-admin users to servers they own list views and config edits enforce ownership, but the...

CVE-2026-79750

MCPHub is a unified hub for centrally managing and dynamically orchestrating multiple MCP servers/APIs into separate endpoints with flexible routing strategies. Prior to version 1.0.30, MCPHub scopes non-admin users to servers they own list views and config edits enforce ownership, but the...

EUVD-2026-68684

MCPHub is a unified hub for centrally managing and dynamically orchestrating multiple MCP servers/APIs into separate endpoints with flexible routing strategies. Prior to version 1.0.30, MCPHub scopes non-admin users to servers they own list views and config edits enforce ownership, but the...

EUVD-2026-68308

Generation of Error Message Containing Sensitive Information vulnerability in ash-project ashai discloses internal error text to chat users. In AshAi.ToolLoop and AshAi.Tools, an exception raised while executing a tool was serialized verbatim with Exception.message/1 into the tool-result content...

CVE-2026-82580 AshAi echoes raw tool exception messages into the conversation, disclosing internal details

Generation of Error Message Containing Sensitive Information vulnerability in ash-project ashai discloses internal error text to chat users. In AshAi.ToolLoop and AshAi.Tools, an exception raised while executing a tool was serialized verbatim with Exception.message/1 into the tool-result content...