PraisonAI is a multi-agent teams system. Prior to 0.1.6, praisonai_platform/services/auth_service.py assigns the public dev-secret-change-me value to JWT_SECRET when PLATFORM_JWT_SECRET is unset, and its production guard does not run when PLATFORM_ENV is also unset because that setting defaults to dev. A remote unauthenticated attacker can mint an HS256 token with an arbitrary sub and email, and the platform's AuthService._verify_token() and get_current_user dependency accept the forged identity for protected API routes. This vulnerability is fixed in praisonai-platform 0.1.6.
PraisonAI versions prior to 0.1.6 contain an improper authentication vulnerability where the system falls back to a public default signing key if the environment variable PLATFORM_JWT_SECRET is unset. This fallback, combined with disabled startup and token-issuance guards in the default dev environment, allows unauthenticated attackers to forge JWT tokens with arbitrary user identities. The vulnerability enables impersonation of users or workspace owners when their identifiers are known. This issue is fixed in version 0.1.6.
PraisonAI is a multi-agent teams system. From 1.5.0 until 1.7.2, MCPServer.startHttp() in src/praisonai-ts/src/mcp/server.ts binds without a host restriction and forwards every HTTP POST request to handleRequest() without authentication or authorization. Any network client that can reach the port can call tools/list, tools/call, resources/read, or prompts/get, causing registered handlers to run with server-side credentials and process privileges or disclose registered data. An initial remediation was released in version 1.7.2.
OpenAM versions prior to 16.1.1 have a vulnerability where the iPlanetDirectoryPro SSO cookie is set without the HttpOnly flag and lacks a protective SameSite attribute. This allows the cookie to be accessed via same-origin cross-site scripting attacks and reused in OAuth and OpenID Connect consent flows, potentially enabling attackers to steal SSO sessions and perform unauthorized consent grants. The issue is resolved in version 16.1.1.
PraisonAI is a multi-agent teams system. Prior to 4.6.58, recipe serve installs APIKeyAuthMiddleware or JWTAuthMiddleware when an operator selects api-key or JWT authentication, but each middleware forwards requests when PRAISONAI_API_KEY or PRAISONAI_JWT_SECRET and the corresponding recipe value are absent. Unauthenticated clients can then reach recipe execution, input, and output surfaces and may trigger connected tools despite the operator explicitly enabling authentication. This issue is fixed in 4.6.58.
NL Portal Backend Libraries provide backend components for Dutch government portals that interact with residents, customers, suppliers, and partner organizations. In versions up to and including 3.0.0, deployments using the shipped default configuration exposed two GraphQL developer features without requiring authentication: the GraphiQL playground, an interactive UI for issuing GraphQL queries; and schema introspection, which lets a caller download the full description of every query, mutation, type, and argument the API supports. Anyone who could reach the `/graphiql` endpoint could open the playground in a browser, pull the full schema, and use that to map out the API and craft calls against it. By itself this does not leak user data, but it removes the guesswork from attacking the rest of the API and significantly lowers the bar for finding and exploiting other weaknesses. Version 3.0.1 patches the issue. As a workaround, override the two settings in deployed configuration.
Certain KGUARD DVR devices running vulnerable firmware expose a system command execution service on all network interfaces without requiring authentication. A remote unauthenticated attacker with network access to the affected service can execute arbitrary system commands on the device, potentially resulting in complete compromise of the DVR.
The vulnerability is known to have been exploited in the wild by the Mirai_ptea (Rimasuta) and Mirai_aurora botnets for malware propagation and subsequent DDoS activity. The vulnerability was reported to affect firmware dating from 2016, while firmware released after 2017 appears to mitigate the issue by restricting the affected service to the localhost interface (127.0.0.1) instead of exposing it on all interfaces (0.0.0.0).
The affected-device list reported by Netlab includes many D1004NR, D1008NR, D1016NR, D1104, D1104NR, D1108NR, D1116NR, D1132NR, D2116NR, D97xx, D98xx, and D99xx variants and several associated hardware revisions
The exploit is included in some version of rapperbot and exploited in 2026. This assignment has been made to document the active exploitation and lack of documentation from the vendor.
Eclipse aeriOS's Identity Manager deployment in its current development version had insecure default configurations and credentials, exposing Keycloak and PostgreSQL services via Kubernetes NodePort and Docker Compose. Default fixed credentials allowed administrative access, risking unauthorized access or modification of identity data. The issue is addressed by randomizing passwords, using Kubernetes Secrets, restricting service exposure, and warning against using predefined development users in production.
Wallos is an open-source, self-hostable personal subscription tracker. Prior to version 5.0.0, the fix for CVE-2026-33407 (GHSA-hhjq-82f8-m6rc, "SSRF via HTTP Proxy Environment Variable") hardened endpoints/logos/search.php by disabling cURL proxying (CURLOPT_PROXY = '' + CURLOPT_NOPROXY = '*'). However, Wallos ships a second, near-identical, unauthenticated logo-image search endpoint — endpoints/payments/search.php — that was not given the same hardening. It still passes the HTTP_PROXY/HTTPS_PROXY environment variable straight into CURLOPT_PROXY. This issue has been patched in version 5.0.0.
oasdiff-action is a GitHub Action that detects breaking changes in OpenAPI specs and post a review on every pull request. Before version 0.0.51, the oasdiff actions resolved external $refs in the OpenAPI spec by default (allow-external-refs: true). When an action runs on a pull request whose spec is attacker-controlled — most importantly fork pull requests on public repositories — a $ref in that spec is fetched/read on the runner with no interaction required, enabling SSRF and disclosure of structured files on the runner. This issue has been patched in version 0.0.51.
CVE-2026-75062 is a critical vulnerability in Google langfun prior to version 0.1.2. It involves improper neutralization of directives in dynamically evaluated code within the default lf.query Python protocol. This flaw allows remote unauthenticated attackers to execute arbitrary Python code by crafting prompt inputs that cause the model to generate executable Python expressions evaluated without sandboxing.
CVE-2026-55581 is an OS command injection vulnerability in sonirico mcp-shell prior to version 0.6.0. The issue arises because the default Docker security.yaml allows /bin/bash in allowed_executables, but validation only checks the first token and does not block the shell command-mode flag -c. This allows an attacker to execute arbitrary commands as the mcpuser. The vulnerability is fixed in version 0.6.0.
Red Hat issued a critical security advisory for Multicluster Global Hub version 1.4.8 addressing multiple vulnerabilities. The update includes security fixes, bug fixes, and updated container images but does not provide explicit patch links or detailed remediation instructions. The advisory references several CVEs with a critical severity rating. Multicluster Global Hub enables management of multiple hub clusters from a single hub cluster.
The Windows installer deployed Npcap leaving its access restriction option at the insecure default value, so the driver was accessible to every local user of the host instead of being restricted to administrators only. A local user without administrative privileges could use the Npcap driver to capture the traffic reaching the host, which discloses information belonging both to the host and to other systems on the same network segment, and to send arbitrary raw packets on that segment.
Deployment of the VPS.org one-click Zulip template deploys a hardcoded application signing key, a default database password ("zulip"), and DISABLE_HTTPS=True.
Deployment of the VPS.org one-click Supabase template deploys a PostgreSQL instance that is published on all interfaces (0.0.0.0:5432) with a default database password set to "postgres". Because Docker installs its own iptables rules, this exposure bypasses a standard host UFW configuration.
Joomla Extension - joomdle.com - Insecure default configuration allows read/write user account access in Joomdle < 3.1.1 - The default configuration of the extension allowed read access and password reset of CMS accounts.
Improper exposure of the MCP server in alibabacloud-rds-openapi-mcp-server allows remote attackers to invoke exposed MCP tools via network access to an MCP endpoint listening on all network interfaces by default.
In NLnet Labs Unbound 1.6.0 up to and including 1.25.1, the 'view_local_data' and 'view_local_datas' commands of 'unbound-control' create a bare local zones tree for an already configured named view when the view is configured with no local data to begin with. However, the creation through the control interface omits adding the default-protected zones (e.g., RFC 1918 reverse, AS112 zones, .onion, .localhost). Once the local zone tree exists without the defaults, every query for a default-protected name from a client mapped to that view escapes to the public DNS via the iterator instead of being answered locally, bypassing local policy expectations.
CVE-2026-47393 affects MervinPraison's PraisonAI multi-agent system. The vulnerability arises because the code-generator for a Flask API server disables authentication by default, exposing critical endpoints without access control. This allows unauthenticated users to interact with the API, potentially accessing sensitive data and controlling LLM orchestration. Versions prior to 4.6.40 ship with this insecure default. The issue is fixed in version 4.6.40 by enabling authentication via configuration.