ghsa

A path traversal vulnerability exists in the XTTS server of the parisneo/lollms package version v9.6. This vulnerability allows an attacker to write audio files to arbitrary locations on the system and enumerate file paths. The issue arises from improper validation of user-provided file paths in the `tts_to_file` endpoint.

Syndic cache directory creation is vulnerable to a directory traversal attack in salt project which can lead a malicious attacker to create an arbitrary directory on a Salt master.

A specially crafted url can be created which leads to a directory traversal in the salt file server. A malicious user can read an arbitrary file from a Salt master’s filesystem.

Parsing a corrupt or malicious image with invalid color indices can cause a panic.

### Impact The implementation of the following functions were determined to include a use-after-free bug: * `FetchEvent.client.tlsCipherOpensslName` * `FetchEvent.client.tlsProtocol` * `FetchEvent.client.tlsClientCertificate` * `FetchEvent.client.tlsJA3MD5` * `FetchEvent.client.tlsClientHello` * `CacheEntry.prototype.userMetadata` of the `fastly:cache` subsystem * `Device.lookup` of the `fastly:device` subsystem This bug could allow for an unintended data leak if the result of the preceding functions were sent anywhere else, and often results in a Compute service crash causing an HTTP 500 error to be returned. As all requests to Compute are isolated from one another, the only data at risk is data present for a single request. ### Patches This bug has been fixed in version 3.16.0 of the `@fastly/js-compute` package. ### Workarounds There are no workarounds for this bug, any use of the affected functions introduces the possibility of a data leak or crash in guest code.

There is a denial-of-service vulnerability in sequoia-openpgp, our crate providing a low-level interface to our OpenPGP implementation. When triggered, the process will enter an infinite loop. Many thanks to Andrew Gallagher for disclosing the issue to us. ## Impact Any software directly or indirectly using the interface `sequoia_openpgp::cert::raw::RawCertParser`. Notably, this includes all software using the `sequoia_cert_store` crate. ## Details The `RawCertParser` does not advance the input stream when encountering unsupported cert (primary key) versions, resulting in an infinite loop. The fix introduces a new raw-cert-specific `cert::raw::Error::UnuspportedCert`. ## Affected software - sequoia-openpgp 1.13.0 - sequoia-openpgp 1.14.0 - sequoia-openpgp 1.15.0 - sequoia-openpgp 1.16.0 - sequoia-openpgp 1.17.0 - sequoia-openpgp 1.18.0 - sequoia-openpgp 1.19.0 - sequoia-openpgp 1.20.0 - Any software built against a vulnerable version of sequoia-openpgp which is directly or i...

### Summary Markdown-based comments in the ZenUML diagram syntax are susceptible to Cross-site Scripting (XSS). ### Details The comment feature allows the user to attach small notes for reference. This feature allows the user to enter in their comment in markdown comment, allowing them to use common markdown features, such as `**` for bolded text. However, the markdown text is currently not sanitized before rendering, allowing an attacker to enter a malicious payload for the comment which leads to XSS. https://github.com/mermaid-js/zenuml-core/blob/dcfee8cde42673c09e19401f43ad8506658c8442/src/components/DiagramFrame/SeqDiagram/MessageLayer/Block/Statement/Comment/Comment.vue#L65 ### PoC ``` // p<img onerror=alert(1) src=""/> A->B:hi ``` Above is a POC diagram payload that results in an XSS. Here is a similar POC in mermaid.live: https://mermaid.live/edit#pako:eNpNjrFuwyAQhl8F3dRK1DaQGhs1kVq1Y6duFQsylwTVgEWw1MTyuxc5S7df39399y0wRIug4IZh9qMOdU2mF-dPJAZMKaa9GTHlB_ZILmnYa9BQH3R4fTq8...

Structs Plugin provides utility functionality used, e.g., in Pipeline to instantiate and configure build steps, typically before their execution. When Structs Plugin 337.v1b_04ea_4df7c8 and earlier fails to configure a build step, it logs a warning message containing diagnostic information that may contain secrets passed as step parameters. This can result in accidental exposure of secrets through the default system log. Structs Plugin 338.v848422169819 inspects the types of actual parameters before logging these warning messages, and limits detailed diagnostic information to FINE level log messages if secrets are involved. These log messages are not displayed in the default Jenkins system log.

When creating secret file credentials Plain Credentials Plugin 182.v468b_97b_9dcb_8 and earlier attempts to decrypt the content of the file to check if it constitutes a valid encrypted secret. In rare cases the file content matches the expected format of an encrypted secret, and the file content will be stored unencrypted (only Base64 encoded) on the Jenkins controller file system. These credentials can be viewed by users with access to the Jenkins controller file system (global credentials) or with Item/Extended Read permission (folder-scoped credentials). Plain Credentials Plugin 183.va_de8f1dd5a_2b_ no longer attempts to decrypt the content of the file when creating secret file credentials.

Bitbucket Branch Source Plugin 886.v44cf5e4ecec5 and earlier prints the Bitbucket OAuth access token as part of the Bitbucket URL in the build log in some cases. Bitbucket Branch Source Plugin 887.va_d359b_3d2d8d does not include the Bitbucket OAuth access token as part of the Bitbucket URL in the build log.