| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| SiYuan before 3.8.10 contains a server-side request forgery vulnerability in the skill-install download path because fetchBytes performs only a parse-time hostname check instead of using ssrfSafeClient. Attackers can supply a skill URL via the MCP skill install action that redirects or DNS-rebinds to loopback addresses, reaching internal services and metadata endpoints. |
| vLLM 0.30.0 before 0.31.0 contains a resource exhaustion vulnerability in OpenCVDynamicOpenPanguVideoBackend, which uses request-controlled num_frames without an effective ceiling. Unauthenticated attackers can send /v1/chat/completions or /tokenize requests selecting the openpangu video_backend with a huge num_frames, forcing np.linspace allocations and CPU work that OOM-kill the API server. |
| vLLM from 0.14.0rc1 before 0.31.0 contains an information disclosure vulnerability in sanitize_message() whose lowercase-only regex fails to scrub uppercase PIL image addresses. Unauthenticated attackers can send a crafted PNG to endpoints such as /tokenize, triggering HF-processor errors that leak live server heap addresses useful for ASLR bypass. |
| vLLM from 0.24.0 before 0.31.0 contains a resource exhaustion vulnerability in the MiniMax M3 VL video processor, which accepts request-level min_pixels without range or min_pixels <= max_pixels validation. Its video-wide pixel budget is enforced only when max_long_side_pixel is set, and that value defaults to None. API clients can supply a large min_pixels via mm_processor_kwargs to enlarge every decoded frame, causing excessive CPU and host-memory consumption that may disrupt workers or the service. |
| vLLM from 0.10.1 before 0.31.0 forwards a request-controlled mm_processor_kwargs.max_patches value to the Hugging Face processor for Cohere2 Vision models without a server-side upper bound, causing quadratic CPU work during multimodal preprocessing. Remote clients can send small image chat-completion requests with large, varying max_patches values to bypass the LRU cache, exhaust preprocessing workers, and deny service to other clients. |
| vLLM from 0.28.0 before 0.30.0 contains a reachable assertion vulnerability in GPUModelRunner._init_xdrope_positions that allows remote attackers to crash the engine by submitting embeddings-only requests. Attackers can send a /v1/completions request containing only prompt_embeds to XD-RoPE models with --enable-prompt-embeds, terminating the EngineCore and API server for all clients. |
| vLLM from 0.19.0 before 0.31.0 contains a resource exhaustion vulnerability that allows inference clients to bypass the server frame limit by setting media_io_kwargs.video.num_frames in chat completion requests. Attackers can supply num_frames=-1 or a large value with many base64 JPEG frames, forcing VideoMediaIO to decode all frames and np.stack them, causing memory exhaustion. |
| Password Pusher from 1.49.0 before 2.17.0 contains an IP spoofing vulnerability that allows remote clients to forge the recorded request IP via unanchored trusted-proxy regular expressions. Clients in 110.0.0.0/8 or 210.0.0.0/8, or connecting directly, can supply X-Forwarded-For values to falsify audit log entries and shift password generator rate limits. |
| Fleet before 4.86.2 contains a denial of service vulnerability that allows unauthenticated attackers to exhaust server memory through quadratic XML canonicalization of unsigned SAML responses. Attackers can send a single crafted SAML response with deeply nested elements declaring new namespace prefixes to POST /api/v1/fleet/sso/callback, crashing the Fleet server. |
| Fleet before 4.89.0 contains an incorrect authorization flaw in its user administration API that accepts user creation or modification when the caller administers any one of the assigned teams. Authenticated team administrators can grant users administrator roles on other teams, then deploy software and execute scripts on those teams' hosts. |
| Coder 2.26.0 before 2.29.24, 2.30.0 before 2.34.16, 2.35.0 before 2.36.11, 2.37.0 before 2.37.8, and 2.38.0 before 2.38.4 contains an improper authorization flaw in the external agent credentials endpoint, which returns the workspace agent token to users with only workspace read access rather than ssh permission. Authenticated users with roles such as Template Admin or Organization Admin can use the token to read the owner's private Git SSH key and Git provider OAuth tokens, and impersonate the agent to receive the owner's sessions. |
| Coder 2.28.0 before 2.29.24, 2.30.0 before 2.34.16, 2.35.0 before 2.36.11, 2.37.0 before 2.37.8, and 2.38.0 before 2.38.4 does not restrict new API keys to the creating token's scopes, allow list, or expiry. A scoped API key or OAuth2 token holding key-creation permission can call POST /api/v2/users/{user}/keys/tokens to mint longer-lived coder:all keys, gaining the user's full privileges. |
| Coder 2.21.0 before 2.29.24, 2.30.0 before 2.34.16, 2.35.0 before 2.36.11, 2.37.0 before 2.37.8, and 2.38.0 before 2.38.4 reads Web Push endpoint responses in webpushSend with io.ReadAll, without a size limit or client timeout. Any authenticated user can register a subscription via POST /api/v2/users/{user}/webpush/subscription pointing to a server returning an endless body, exhausting coderd memory and crashing the API, dashboard, and workspace connections. |
| The wasmtime-wasi-http p2 implementation in Wasmtime before 36.0.20, 37.0.0 before 48.0.8, 49.0.0 before 49.0.5, and 50.0.0 before 50.0.1 can send conflicting content-length headers, or ones not matching the body, in outgoing requests. A guest that reuses a received response's headers resource to send a bodyless request can enable request smuggling against non-compliant servers, potentially bypassing embedder destination access controls. |
| Wasmtime before 36.0.20, 37.0.0 before 48.0.8, 49.0.0 before 49.0.5, and 50.0.0 before 50.0.1 allocates untracked host memory when a component receives a borrow of a resource it defines itself, bypassing the per-store handle limit. A malicious guest component can repeatedly transfer such borrowed resources to consume arbitrary host memory, causing denial of service. |
| Wasmtime 47.0.0 before 48.0.8, 49.0.0 before 49.0.5, and 50.0.0 before 50.0.1 contains an insufficient type-check in the TypedFunc embedder API for I31, Option<I31>, and Option<NoneRef> parameter or result types, reaching an unreachable!() path. A malicious guest module exporting a function that uses another anyref-hierarchy type can panic the host embedding, causing denial of service. |
| Wasmtime 47.0.0 before 48.0.8, 49.0.0 before 49.0.5, and 50.0.0 before 50.0.1 contains a cross-engine type confusion vulnerability because StructType::with_finality_and_supertype does not validate that the supplied supertype belongs to the same Engine. Embedders using multiple engines that create a GC struct type with another engine's supertype can trigger runtime panics or GC heap corruption, causing denial of service; guest WebAssembly cannot trigger it. |
| Wasmtime's wasmtime-wasi crate 48.0.0 before 48.0.8, 49.0.0 before 49.0.5, and 50.0.0 before 50.0.1 contains a time-of-check/time-of-use race in the wasip3 wasi:sockets udp-socket host implementation that lets guests bypass socket_addr_check send restrictions. A malicious guest can race udp-socket.send against udp-socket.connect to deliver attacker-controlled datagrams to addresses the policy permits only for receiving. |
| Frappe HRMS 16.0.0 before 16.50.0 lacks authorization checks on certain HR workflow endpoints, letting authenticated users access records they are not permitted to view. Any logged-in user, including one with only the Employee role, can read candidate record details and employee email addresses, and cause emails to be sent to employees. |
| Frappe HRMS 15.0.0 before 15.64.1 and 16.0.0 before 16.19.0 lacks a write permission check on a bulk record processing action, allowing users with only read access to the related setup record to modify data. Authenticated users with the default Employee role can run the action to create and submit records for every covered employee. |
