Understand vulnerabilities. Protect your applications.
Discover which vulnerabilities affect your technologies and how to fix them. Explore clear explanations, affected versions and practical recommendations - from CMS platforms and plugins to application libraries.
Vulnerability archive 456
Reset filtersvirtualenv: Deletion recovery can change a symlink target's mode
virtualenv: Builtin creation follows linked environment directories
unstructured: Stored XSS via unescaped HTML generation
Pillow: P;2L and P;4L decoders may read 1 to 3 bytes past the passed buffer size
An out-of-bounds read vulnerability in Pillow allows certain decoders to read heap memory past the end of a buffer when processing specific image dimensions. This may disclose adjacent memory into decoded pixel data or cause an application crash.
Pillow: NULL dereference in ImagingExpand via rank filters on zero-size images
Applying rank, median, minimum, or maximum filters to zero-width or zero-height images in Pillow triggers a NULL pointer dereference in ImagingExpand, causing Python worker processes to crash and enabling denial of service.
Pillow: Argument Injection in Ghostscript call, potentially leading into RCE
Pillow lacks sanitization when constructing Ghostscript commands from input filenames, permitting command-line argument injection that can be escalated to remote code execution.
Pillow: Heap buffer overflow in ImagingJpegDecode
A heap buffer overflow vulnerability in Pillow allows attackers to cause out-of-bounds writes when decoding specially crafted lossless JPEG images using draft scaling. This issue occurs when libjpeg-turbo silently ignores scaling requests, leading to memory corruption or crashes.
Pillow: Unbounded recursion parsing a PDF page tree or trailer chain (incomplete fix for CVE-2026-42310)
Unbounded recursion in Pillow when parsing PDF page trees or trailer chains can cause severe CPU and memory exhaustion or crashes during PDF append operations.
certifi bundles emSign root CAs with compromised intermediate CA keys
Certain certifi releases bundled a root certificate with compromised intermediate keys. An attacker on the network path could issue forged certificates for any hostname and intercept or modify TLS connections.
Pipecat: Public exposure of Pipecat’s development runner can allow unauthenticated bot session creation
Exposing Pipecat's development runner to untrusted networks allows unauthenticated callers to create bot sessions and paid Daily rooms. This can lead to unauthorized bot interactions, local resource exhaustion, and consumption of third-party service quotas.
ClickMigrate Path Traversal in "revision -m" Argument
A path traversal vulnerability in ClickMigrate allows an attacker who controls the migration name argument to create a file in an arbitrary writable location outside the intended migrations directory.
Apprise forwards configured auth headers across cross-origin HTTP redirects
Apprise HTTP notification plugins and loaders follow redirects by default while re-sending configured authentication headers. If a configured destination is compromised or subjected to a man-in-the-middle attack, secrets can be leaked to an arbitrary cross-origin host.
Shiny for Python has path traversal in bookmark restore
A path traversal vulnerability in Shiny for Python allows unauthenticated remote attackers to probe file existence and JSON validity via bookmark restoration, or read arbitrary files when server-side bookmarking and file inputs are configured.
pyLoad: Api.set_user_permission never invalidates the target's session
pyLoad WindowsPhoneNotify addon: non-admin SETTINGS user triggers SSRF via unguarded http.client notification host
pyLoad: Tar extraction creates device nodes and FIFOs (member types not filtered; tarfile extractall without filter=)
Strawberry legacy graphql-ws retains naturally completed subscription slots
PraisonAI: AICoder Arbitrary File Write and Command Execution via LLM Tool Calls
datamodel-code-generator: Protobuf weak-import path traversal allows files to be written outside the temporary directory
PraisonAI: Project custom command templates can read outside-workspace files into model prompts
Pydantic AI OpenTelemetry instrumentation: retry prompt content is not redacted when `include_content=False`
Pydantic AI OpenTelemetry instrumentation: exception events on tool and agent run spans include content when `include_content=False`
PraisonAI: Prompt Injection Defense Bypassed for HIGH-Level Threats
Know which vulnerabilities affect your application
With the DockRay integration and software inventory enabled, DockRay checks reported component versions against known vulnerabilities and notifies you about confirmed matches. Coverage depends on the supported technology and available advisory data.