Out-of-bounds Write
The product writes data past the end, or before the beginning, of the intended buffer.
Software weakness types referenced by NVD CVE records. Includes the MITRE Top 25 most dangerous weaknesses and other commonly seen entries.
The product writes data past the end, or before the beginning, of the intended buffer.
Improper neutralization of input during web page generation lets attackers inject client-side scripts.
User-controlled input is concatenated into SQL queries, allowing attackers to alter query semantics.
Referencing memory after it has been freed can lead to crashes, data corruption, or arbitrary code execution.
User input is incorporated into a command passed to a shell or OS, enabling command execution.
Input is not validated or is incorrectly validated, leading to unexpected control or data flow.
The product reads data past the end, or before the beginning, of the intended buffer.
User input is used to construct a pathname that resolves outside the intended directory.
The product does not verify that requests came from the user, allowing forged actions.
Files of dangerous types are uploaded without proper validation, leading to remote execution.
Sensitive functionality lacks authorization checks, letting unauthenticated users perform privileged actions.
A pointer expected to be valid is NULL when dereferenced, typically causing a crash.
The product does not properly verify the identity of an actor, enabling impersonation.
Arithmetic operations produce a value outside the valid range, leading to incorrect logic.
Untrusted serialized data is deserialized, often leading to remote code execution.
User input is included in a constructed command, allowing attackers to modify the command.
Operations on a buffer can read or write outside its bounds.
Hard-coded credentials embedded in code create a backdoor for attackers.
A server fetches a URL on behalf of a user, allowing access to internal resources.
Critical functions are exposed without requiring authentication.
Concurrent code paths can interleave in ways that produce incorrect or insecure behavior.
Privileges are assigned or maintained incorrectly, enabling privilege escalation.
User-controlled input is interpreted as code, allowing arbitrary code execution.
Authorization is performed but does not correctly enforce the intended policy.
Default file or resource permissions are too permissive, exposing data or capabilities.
Sensitive information is exposed to an actor not explicitly authorized.
The product does not properly restrict resource use, enabling denial of service.
User input is used to construct a redirect URL, enabling phishing.
An XML parser is configured to resolve external entities, exposing files or SSRF.
A critical resource has permissions that allow unintended access.
A temporary file is created in a way that allows attackers to predict or hijack it.
Sensitive data is not encrypted at rest or in transit.
Certificates are not validated, enabling MITM attacks.
Predictable random values are used in security contexts.
Resources are loaded from a path that an attacker can control.
Spotted something off, or have an idea? Let us know.