Palantir Gotham included an endpoint that would log arbitrary sized payloads.

Palantir Gotham versions prior to 3.22.11.2 included an unauthenticated endpoint that would have allowed an attacker to exhaust the memory of the Gotham dispatch service.

Authentication bypass in Aries due to misconfiguration

A vulnerability in Palantir's Aries service allowed unauthenticated access to log viewing and management functionality on Apollo instances using default configuration. The defect resulted in both authentication and authorization checks being bypassed, potentially allowing any network-accessible client to view system logs and perform operations without valid credentials. No evidence of exploitation was identified during the vulnerability window.

Gaia unauthenticated endpoints

Gotham Gaia application was found to be exposing multiple unauthenticated endpoints.

Insufficient permission checks when pre-enrolling users Summary

Control Panel provides an API for pre-registering into an enrollment and organization prior to a user's first login. The API for creating users checks that the account requesting a user creation has `edit` on the enrollment-level user directory, but is missing a separate check that the enrollment editor has access (or belongs to) the organization that they are adding a user to.

Glutton V1 endpoints missing authentication

Glutton V1 service endpoints were exposed without any authentication on Gotham stacks, this could have allowed users that did not have any permission to hit glutton backend directly and read/update/delete data. The affected service has been patched and automatically deployed to all Apollo-managed Gotham Instances

Network boundaries not respected in certain Foundry namespaces.

Due to a product misconfiguration in certain deployment types, it was possible from different pods in the same namespace to communicate with each other. This issue resulted in bypass of access control due to the presence of a vulnerable endpoint in Foundry Container Service that executed user-controlled commands locally.

Access control issue impacting RV backed objects

Restricted Views backed objects (OSV1) could be bypassed under specific circumstances due to a software bug, this could have allowed users that didn't have permission to see such objects to view them via Object Explorer directly. This software bug did not impact or otherwise make data available across organizational boundaries nor did it allow for data to be viewed or accessed by unauthenticated users. The affected service have been patched and automatically deployed to all Apollo-managed Foundry instances.

Gotham table and Forward App Path traversal

Gotham Table service and Forward App were found to be vulnerable to a Path traversal issue allowing an authenticated user to read arbitrary files on the file system.

The Foundry Blobster service was found to have a cross-site scripting (XSS) vulnerability.

The Foundry Blobster service was found to have a cross-site scripting (XSS) vulnerability that could have allowed an attacker with access to Foundry to launch attacks against other users. This vulnerability is resolved in Blobster 3.228.0.

The Foundry Code-Workbooks service was found to contain an issue leading to information disclosure.

Information Exposure Through Log Files vulnerability discovered in Foundry Code-Workbooks where the endpoint backing that console was generating service log records of any Python code being run. These service logs included the Foundry token that represents the Code-Workbooks Python console. Upgrade to Code-Workbooks version 4.461.0. This issue affects Palantir Foundry Code-Workbooks version 4.144 to version 4.460.0 and is resolved in 4.461.0.

A component in Foundry logging was found to be capturing sensitive information in logs.

Information Exposure Through Log Files vulnerability discovered in Foundry when logs were captured using an underlying library known as Build2. This issue was present in versions earlier than 1.785.0. Upgrade to Build2 version 1.785.0 or greater.

Palantir Gotham included an unauthenticated endpoint that listed all active usernames in the platform with an active session.

Palantir Gotham included an unauthenticated endpoint that listed all active usernames on the stack with an active session. The affected services have been patched and automatically deployed to all Apollo-managed Gotham instances. It is highly recommended that customers upgrade all affected services to the latest version. This issue affects: Palantir Gotham versions prior to 103.30221005.0.

It was discovered that the sls-logging was not verifying hostnames in TLS certificates due to a misuse of the javax.net.ssl.SSLSocketFactory API. A malicious attacker in a privileged network position could abuse this to perform a man-in-the-middle attack. A successful man-in-the-middle attack would allow them to intercept, read, or modify network communications to and from the affected service. In the case of AtlasDB, the vulnerability was mitigated by other network controls such as two-way TLS when deployed as part of a Palantir platform. Palantir still recommends upgrading to a non-vulnerable version out of an abundance of caution.

Palantir Gotham included an endpoint that would log arbitrary sized zip files.

Palantir Gotham versions prior to 3.22.11.2 included an unauthenticated endpoint that would load portions of maliciously crafted zip files to memory. An attacker could repeatedly upload a malicious zip file, which would allow them to exhaust memory resources on the dispatch server.

Gotham Chat IRC help does not validate hostnames in TLS certificates

Improper Validation of Certificate with Host Mismatch vulnerability in Gotham Chat IRC helper of Palantir Gotham allows A malicious attacker in a privileged network position could abuse this to perform a man-in-the-middle attack. A successful man-in-the-middle attack would allow them to intercept, read, or modify network communications to and from the affected service. This issue affects: Palantir Palantir Gotham Chat IRC helper versions prior to 30221005.210011.9242.

It was discovered that the Magritte-ftp was not verifying hostnames in TLS certificates due to a misuse of the javax.net.ssl.SSLSocketFactory API. A malicious attacker in a privileged network position could abuse this to perform a man-in-the-middle attack. A successful man-in-the-middle attack would allow them to intercept, read, or modify network communications to and from the affected service. In the case of a successful man in the middle attack on magritte-ftp, an attacker would be able to read and modify network traffic such as authentication tokens or raw data entering a Palantir Foundry stack.

It was discovered that the sls-logging was not verifying hostnames in TLS certificates due to a misuse of the javax.net.ssl.SSLSocketFactory API. A malicious attacker in a privileged network position could abuse this to perform a man-in-the-middle attack. A successful man-in-the-middle attack would allow them to intercept, read, or modify network communications to and from the affected service.

Retrieval of Attachments to Comments lacks Authorization

A security defect in Foundry's Comments functionality resulted in the retrieval of attachments to comments not being gated by additional authorization checks. This could enable an authenticated user to inject a prior discovered attachment UUID into other arbitrary comments to discover it's content. This defect was fixed in Foundry Comments 2.249.0, and a patch was rolled out to affected Foundry environments. No further intervention is required at this time.

Mandatory control bypass in Lime2

Palantir Foundry deployments running Lime2 versions between 2.519.0 and 2.532.0 were vulnerable a bug that allowed authenticated users within a Foundry organization to bypass discretionary or mandatory access controls under certain circumstances.

CVE-2023-30945

Multiple Services such as VHS(Video History Server) and VCD(Video Clip Distributor) and Clips2 were discovered to be vulnerable to an unauthenticated arbitrary file read/write vulnerability due to missing input validation on filenames. A malicious attacker could read sensitive files from the filesystem or write/delete arbitrary files on the filesystem as well.