CVE Vulnerability Catalog

databasement before 1.8.2 allows remote code execution because it runs certain commands (e.g., mariadb-dump) with a database name that can be specified by any authenticated user. For example, --result-file=/app/public/index.php can write to index.php. In other words, quoting prevents OS command injection in mariadb-dump, but the argument injection alone is sufficient for code execution indirectly.

A vulnerability in the Zephyr Bluetooth LE stack allows a remote, unauthenticated attacker to trigger a use-after-free in L2CAP CoC channel handling with dynamic PSMs. By sending a data frame followed by a disconnect request, the attacker can cause the system workqueue to process work referencing a freed channel, leading to memory corruption or workqueue corruption.

The NXP MCUX TRNG entropy driver in Zephyr (drivers/entropy/entropy_mcux_trng.c) passes the caller's byte count directly to the SDK routine TRNG_GetRandomData(), which on i.MX RT5xx/RT6xx parts always copies whole 32-bit words and rounds entropy draws up to a multiple of 128 bytes. A request length not a multiple of four causes an underflow in the SDK's buffer-full check, leading to an overflow of up to 127 bytes past the caller's buffer.

A vulnerability in Zephyr's syscall verifiers z_vrfy_fuel_gauge_get_props() and z_vrfy_fuel_gauge_set_props() allocates variable-length arrays on the kernel stack sized directly by an unvalidated caller-supplied len argument. Because no bounds check occurs before allocation and copying, a user-mode thread (with CONFIG_USERSPACE) can trigger an out-of-bounds write in supervisor mode with attacker-controlled length and, on the permitted path, attacker-controlled content.

An out-of-bounds read vulnerability was found in the IPv6 subsystem of the Zephyr stack in net_route_ipv6_packet(). The issue stems from missing validation of whether a neighbor cache entry has a link-layer address before using it. An attacker on the same link can force an unresolved neighbor state and trigger an out-of-bounds read past the address array, potentially causing a system panic or data leak.

omarchy-theme-set in Omarchy 4 before 4.0.1 allows code execution via a third-party theme because the files placed into ~/.local/state/omarchy/current/theme may include executable content from an untrusted Git repository.

pH7Builder (pH7 Social Dating CMS) before 18.6.0 contains a hard-coded API key vulnerability in Tool.class.php that allows unauthenticated attackers to bypass API access checks by spoofing the Host header. Attackers can send Host: localhost with private_api_key=dev772277 and the default allowed URL to retrieve member emails, IP addresses, phone numbers, and bank account fields.

pH7Builder (pH7 Social Dating CMS) before 18.5.0 contains a path traversal vulnerability in the picture module deletePhoto() action that allows authenticated members to delete arbitrary files. Attackers can supply ../ sequences in the POST picture_link parameter to remove other members' photos or configuration and cache files, causing content loss and denial of service.

AmoyLab Unla through 0.10.0 contains an authentication bypass vulnerability that allows unauthenticated attackers to obtain valid access tokens because the OAuth2 server never authenticates a resource owner. Attackers can register a client, request a code from /authorize, and exchange it at /token to access OAuth2-protected MCP prefixes, proxied upstream APIs and injected credentials.

Katanemo Plano through 0.4.37 contains a missing authentication vulnerability that allows unauthenticated network attackers to access the Envoy admin interface, which is bound to all host interfaces on port 9901. Attackers can request the /config_dump endpoint to read configured LLM provider API keys in plaintext from the WASM filter configuration.

mcp-go through 1.2.1 contains a denial of service vulnerability in StreamableHTTPServer.ServeHTTP that allows remote unauthenticated attackers to exhaust memory by sending oversized POST bodies. Attackers can send arbitrarily large or many concurrent POST requests, read fully via io.ReadAll before validation, to degrade or OOM-kill the server process.

UnicomAI Wanwu before 0.6.3 contains an insecure direct object reference vulnerability that allows authenticated low-privileged users to delete other tenants' agent or RAG applications by supplying their appId. Attackers can send requests to DELETE /v1/appspace/app with guessed sequential assistant IDs to permanently delete victims' applications, workflows, conversations, and associated data.

LlamaFarm through 0.0.34 contains an insecure default configuration that binds its unauthenticated FastAPI server to 0.0.0.0 on port 14345, while the lf CLI silently discards HOST overrides. Network-adjacent attackers can call the project and dataset management API to read stored provider API keys, modify projects, trigger ingestion, and irreversibly delete projects.

Easy!Appointments through 1.6.0 contains an authorization bypass vulnerability in Booking::register() that allows unauthenticated attackers to modify any appointment by supplying an appointment id without its hash. Attackers can enumerate sequential appointment ids with a self-asserted manage_mode flag to rewrite appointment details, rebind them to attacker-controlled customers, and obtain management hashes for rescheduling or cancellation.

Vearch 3.5.2 through 3.5.9 contains an incorrect authorization vulnerability in Role.HasPermissionForResources that ignores stored ReadOnly or None privilege levels for resources listed in a role. Authenticated non-root users can upsert and delete documents with read-only access, or call PUT /roles to grant their role WriteRead privileges, escalating toward cluster administrator access.

pbi-cli 3.10.1 through 3.12.0 contains an OS command injection vulnerability in desktop_sync.py that passes unquoted .pbip paths to cmd /c start when reopening projects. Attackers can lure victims into opening a Power BI project from a space-free path containing & to run commands with victim privileges during report write or reload.

GoatCounter through 2.7.0 contains a mass assignment privilege escalation vulnerability in the userPrefSave handler that allows logged-in users to modify protected account fields via form-encoded requests. Attackers with read-only access can POST user.access[all]=* and user.email_verified=true to /user/pref, bypassing readonly tags to gain superuser or admin access.

OpenAgents Workspace backend through launcher-v1.0.17 contains an information disclosure vulnerability that allows unauthenticated attackers to list all workspaces via GET /v1/workspaces. Attackers can read the unmasked browserfabric_api_key in each workspace's settings map, along with workspace ids, slugs, creator emails and member lists.

Rill 0.77.0 through 0.90.5 contains a missing authorization vulnerability in the admin OAuth server that issues authorization codes to dynamically registered clients without user consent. Attackers can register a client with the long_lived_access_token scope and lure a user to an authorization link, obtaining a non-expiring API token with the user's full permissions.

MCP Kotlin SDK through 0.15.0 contains an uncontrolled memory allocation vulnerability that allows remote clients to exhaust server memory because Application.mcpWebSocket installs Ktor WebSockets without a maxFrameSize limit. Attackers can send small frame headers declaring payloads near 2 GiB over one or a few connections, forcing huge heap allocations and causing denial of service.