CVEs

Graphiti is a framework for building and querying temporal context graphs for AI agents. 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. In MCP deployments, this was exploitable not only through direct untrusted access to the Graphiti MCP server, but also thro

Tinyauth is an authentication and authorization server. Prior to 5.0.3, the OIDC authorization endpoint allows users with a TOTP-pending session (password verified, TOTP not yet completed) to obtain authorization codes. An attacker who knows a user's password but not their TOTP secret can obtain valid OIDC tokens, completely bypassing the second factor. This vulnerability is fixed in 5.0.3.

Parse Server is an open source backend that can be deployed to any infrastructure that can run Node.js. Prior to 9.6.0-alpha.11 and 8.6.37, Parse Server's built-in OAuth2 auth adapter exports a singleton instance that is reused directly across all OAuth2 provider configurations. Under concurrent authentication requests for different OAuth2 providers, one provider's token validation may execute using another provider's configuration, potentially allowing a token that should be rejected by one pro

Backstage is an open framework for building developer portals. Prior to 0.27.1, a Server-Side Request Forgery (SSRF) vulnerability exists in @backstage/plugin-auth-backend when auth.experimentalClientIdMetadataDocuments.enabled is set to true. The CIMD metadata fetch validates the initial client_id hostname against private IP ranges but does not apply the same validation after HTTP redirects. The practical impact is limited. The attacker cannot read the response body from the internal request, c

ZeptoClaw is a personal AI assistant. Prior to 0.7.6, the generic webhook channel trusts caller-supplied identity fields (sender, chat_id) from the request body and applies authorization checks to those untrusted values. Because authentication is optional and defaults to disabled (auth_token: None), an attacker who can reach POST /webhook can spoof an allowlisted sender and choose arbitrary chat_id values, enabling high-risk message spoofing and potential IDOR-style session/chat routing abuse. T

NEXULEAN is a cybersecurity portfolio & service platform for an Ethical Hacker, AI Enthusiast, and Penetration Tester. Prior to 2.0.0, a security vulnerability was identified where Firebase and Web3Forms API keys were exposed. An attacker could use these keys to interact with backend services without authentication, potentially leading to unauthorized access to application resources and user data. This vulnerability is fixed in 2.0.0.

An information disclosure issue in the zipfileInflate function in the zipfile extension in SQLite v3.51.1 and earlier allows attackers to obtain heap memory via supplying a crafted ZIP file.

A command injection vulnerability has been identified in the Telnet command-line interface (CLI) of TP-Link TL-MR6400 v5.3. This issue is caused by insufficient sanitization of data processed during specific CLI operations. An authenticated attacker with elevated privileges may be able to execute arbitrary system commands. Successful exploitation may lead to full device compromise, including potential loss of confidentiality, integrity, and availability.

Dataease is an open source data visualization analysis tool. Prior to 2.10.20, By controlling the IniFile parameter, an attacker can force the JDBC driver to load an attacker-controlled configuration file. This configuration file can inject dangerous JDBC properties, leading to remote code execution. The Redshift JDBC driver execution flow reaches a method named getJdbcIniFile. The getJdbcIniFile method implements an aggressive automatic configuration file discovery mechanism. If not explicitly

Dataease is an open source data visualization analysis tool. Prior to 2.10.20, The table parameter for /de2api/datasource/previewData is directly concatenated into the SQL statement without any filtering or parameterization. Since tableName is a user-controllable string, attackers can inject malicious SQL statements by constructing malicious table names. This vulnerability is fixed in 2.10.20.

Magic Wormhole makes it possible to get arbitrary-sized files and directories from one computer to another. From 0.21.0 to before 0.23.0, receiving a file (wormhole receive) from a malicious party could result in overwriting critical local files, including ~/.ssh/authorized_keys and .bashrc. This could be used to compromise the receiver's computer. Only the sender of the file (the party who runs wormhole send) can mount the attack. Other parties (including the transit/relay servers) are excluded

A Missing Authorization vulnerability in Trane Tracer SC, Tracer SC+, and Tracer Concierge could allow an unauthenticated attacker to access sensitive information through unprotected APIs.

A Memory Allocation with Excessive Size Value vulnerability in Trane Tracer SC, Tracer SC+, and Tracer Concierge could allow an unauthenticated attacker to cause a denial-of-service condition

GL-iNet GL-AR300M16 v4.3.11 was discovered to contain a SQL injection vulnerability via the add_group() function. This vulnerability allows attackers to execute arbitrary SQL database operations via a crafted HTTP request.

Tina is a headless content management system. Prior to 2.1.8, the TinaCMS CLI development server exposes media endpoints that are vulnerable to path traversal, allowing attackers to read and write arbitrary files on the filesystem outside the intended media directory. When running tinacms dev, the CLI starts a local HTTP server (default port 4001) exposing endpoints such as /media/list/*, /media/upload/*, and /media/*. These endpoints process user-controlled path segments using decodeURI() and p

Tina is a headless content management system. Prior to 2.1.7, a path traversal vulnerability exists in the TinaCMS development server's media upload handler. The code at media.ts joins user-controlled path segments using path.join() without validating that the resulting path stays within the intended media directory. This allows writing files to arbitrary locations on the filesystem. This vulnerability is fixed in 2.1.7.

llama.cpp is an inference of several LLM models in C/C++. Prior to b8146, the gguf_init_from_file_impl() in gguf.cpp is vulnerable to an Integer overflow, leading to an undersized heap allocation. Using the subsequent fread() writes 528+ bytes of attacker-controlled data past the buffer boundary. This is a bypass of a similar bug in the same file - CVE-2025-53630, but the fix overlooked some areas. This vulnerability is fixed in b8146.

Postal is an open source SMTP server. Postal versions less than 3.3.5 had a HTML injection vulnerability that allowed unescaped data to be included in the admin interface. The primary way for unescaped data to be added is via the API's "send/raw" method. This could allow arbitrary HTML to be injected in to the page which may modify the page in a misleading way or allow for unauthorised javascript to be executed. Fixed in 3.3.5 and higher.