A vulnerability in danswer-ai/danswer version 1 allows an attacker to perform a Regular Expression Denial of Service (ReDoS) by manipulating regular expressions. This can significantly slow down the application's response time and potentially render it completely unusable.

Vanna-ai v0.6.2 is vulnerable to SQL Injection due to insufficient protection against injecting additional SQL commands from user requests. The vulnerability occurs when the `generate_sql` function calls `extract_sql` with the LLM response. An attacker can include a semi-colon between a search data field and their own command, causing the `extract_sql` function to remove all LLM generated SQL and execute the attacker's command if it passes the `is_sql_valid` function. This allows the execution of user-defined SQL beyond the expected boundaries, notably the trained schema.

A remote code execution vulnerability exists in the Calculate function of parisneo/lollms version 9.8. The vulnerability arises from the use of Python's `eval()` function to evaluate mathematical expressions within a Python sandbox that disables `__builtins__` and only allows functions from the `math` module. This sandbox can be bypassed by loading the `os` module using the `_frozen_importlib.BuiltinImporter` class, allowing an attacker to execute arbitrary commands on the server. The issue is fixed in version 9.10.

A Cross-Site Request Forgery (CSRF) vulnerability exists in the latest commit (56b782bcefd2e59b19cd7ba7878b95f54884f502) of the vanna-ai/vanna repository. Two endpoints in the built-in web app that provide SQL functionality are implemented as simple GET requests, making them susceptible to CSRF attacks. This vulnerability allows an attacker to run arbitrary SQL commands via CSRF without the target intending to expose the web app to the network or other users. The impact is limited to data alteration or deletion, as the attacker cannot read the results of the query.

Gunicorn version 21.2.0 does not properly validate the value of the 'Transfer-Encoding' header as specified in the RFC standards, which leads to the default fallback method of 'Content-Length,' making it vulnerable to TE.CL request smuggling. This vulnerability can lead to cache poisoning, data exposure, session manipulation, SSRF, XSS, DoS, data integrity compromise, security bypass, information leakage, and business logic abuse.

In the latest version of pytorch/serve, the script 'upload_results_to_s3.sh' references the S3 bucket 'benchmarkai-metrics-prod' without ensuring its ownership or confirming its accessibility. This could lead to potential security vulnerabilities or unauthorized access to the bucket if it is not properly secured or claimed by the appropriate entity. The issue may result in data breaches, exposure of proprietary information, or unauthorized modifications to stored data.

A path traversal vulnerability exists in stitionai/devika, specifically in the project creation functionality. In the affected version beacf6edaa205a5a5370525407a6db45137873b3, the project name is not validated, allowing an attacker to create a project with a crafted name that traverses directories. This can lead to arbitrary file overwrite when the application generates code and saves it to the specified project directory, potentially resulting in remote code execution.

Due to a lack of access control, unauthorized users are able to view and modify information pertaining to other users.

An Out-Of-Memory (OOM) vulnerability exists in the `ollama` server version 0.3.14. This vulnerability can be triggered when a malicious API server responds with a gzip bomb HTTP response, leading to the `ollama` server crashing. The vulnerability is present in the `makeRequestWithRetry` and `getAuthorizationToken` functions, which use `io.ReadAll` to read the response body. This can result in excessive memory usage and a Denial of Service (DoS) condition.

A stored cross-site scripting (XSS) vulnerability exists in infiniflow/ragflow, affecting the latest commit on the main branch (cec2080). The vulnerability allows an attacker to upload HTML/XML files that can host arbitrary JavaScript payloads. These files are served with the 'application/xml' content type, which is automatically rendered by browsers. This can lead to the execution of arbitrary JavaScript in the context of the user's browser, potentially allowing attackers to steal cookies and gain unauthorized access to user files and resources. The vulnerability does not require authentication, making it accessible to anyone with network access to the instance.

A Denial of Service (DoS) vulnerability exists in the brycedrennan/imaginairy repository, version 15.0.0. The vulnerability is present in the `/api/stablestudio/generate` endpoint, which can be exploited by sending an invalid request. This causes the server process to terminate abruptly, outputting `KILLED` in the terminal, and results in the unavailability of the server. This issue disrupts the server's functionality, affecting all users.

A vulnerability in the gradio-app/gradio repository, version git 67e4044, allows for path traversal on Windows OS. The implementation of the blocked_path functionality, which is intended to disallow users from reading certain files, is flawed. Specifically, while the application correctly blocks access to paths like 'C:/tmp/secret.txt', it fails to block access when using NTFS Alternate Data Streams (ADS) syntax, such as 'C:/tmp/secret.txt::$DATA'. This flaw can lead to unauthorized reading of blocked file paths.

A vulnerability in the `ImageClassificationDataset.from_csv()` API of the `dmlc/gluon-cv` repository, version 0.10.0, allows for arbitrary file write. The function downloads and extracts `tar.gz` files from URLs without proper sanitization, making it susceptible to a TarSlip vulnerability. Attackers can exploit this by crafting malicious tar files that, when extracted, can overwrite files on the victim's system via path traversal or faked symlinks.

In kedro-org/kedro version 0.19.8, the `pull_package()` API function allows users to download and extract micro packages from the Internet. However, the function `project_wheel_metadata()` within the code path can execute the `setup.py` file inside the tar file, leading to remote code execution (RCE) by running arbitrary commands on the victim's machine.

A remote code execution vulnerability exists in open-mmlab/mmdetection version v3.3.0. The vulnerability is due to the use of the `pickle.loads()` function in the `all_reduce_dict()` distributed training API without proper sanitization. This allows an attacker to execute arbitrary code by broadcasting a malicious payload to the distributed training network.

A remote code execution vulnerability exists in invoke-ai/invokeai versions 5.3.1 through 5.4.2 via the /api/v2/models/install API. The vulnerability arises from unsafe deserialization of model files using torch.load without proper validation. Attackers can exploit this by embedding malicious code in model files, which is executed upon loading. This issue is fixed in version 5.4.3.

A Cross-Origin Resource Sharing (CORS) vulnerability exists in feast-dev/feast version 0.40.0. The CORS configuration on the agentscope server does not properly restrict access to only trusted origins, allowing any external domain to make requests to the API. This can bypass intended security controls and potentially expose sensitive information.

A stored cross-site scripting (XSS) vulnerability exists in Serge version 0.9.0. The vulnerability is due to improper neutralization of input during web page generation in the chat prompt. An attacker can exploit this vulnerability by sending a crafted message containing malicious HTML/JavaScript code, which will be stored and executed whenever the chat is accessed, leading to unintended content being shown to the user and potential phishing attacks.

A missing check_access() function in the lollms_binding_infos module of the parisneo/lollms repository, version V14, allows attackers to add, modify, and remove bindings arbitrarily. This vulnerability affects the /install_binding and /reinstall_binding endpoints, among others, enabling unauthorized access and manipulation of binding settings without requiring the client_id value.

A Denial of Service (DoS) vulnerability was discovered in the /api/v1/boards/{board_id} endpoint of invoke-ai/invokeai version v5.0.2. This vulnerability occurs when an excessively large payload is sent in the board_name field during a PATCH request. By sending a large payload, the UI becomes unresponsive, rendering it impossible for users to interact with or manage the affected board. Additionally, the option to delete the board becomes inaccessible, amplifying the severity of the issue.

In invoke-ai/invokeai version v5.0.2, the web API `POST /api/v1/images/delete` is vulnerable to Arbitrary File Deletion. This vulnerability allows unauthorized attackers to delete arbitrary files on the server, potentially including critical or sensitive system files such as SSH keys, SQLite databases, and configuration files. This can impact the integrity and availability of applications relying on these files.

A vulnerability in langchain-core versions >=0.1.17,<0.1.53, >=0.2.0,<0.2.43, and >=0.3.0,<0.3.15 allows unauthorized users to read arbitrary files from the host file system. The issue arises from the ability to create langchain_core.prompts.ImagePromptTemplate's (and by extension langchain_core.prompts.ChatPromptTemplate's) with input variables that can read any user-specified path from the server file system. If the outputs of these prompt templates are exposed to the user, either directly or through downstream model outputs, it can lead to the exposure of sensitive information.

A Denial of Service (DoS) vulnerability in the multipart request boundary processing mechanism of the Invoke-AI server (version v5.0.1) allows unauthenticated attackers to cause excessive resource consumption. The server fails to handle excessive characters appended to the end of multipart boundaries, leading to an infinite loop and a complete denial of service for all users. The affected endpoint is `/api/v1/images/upload`.

A vulnerability in szad670401/hyperlpr v3.0 allows for a Denial of Service (DoS) attack. The server fails to handle excessive characters appended to the end of multipart boundaries, regardless of the character used. This flaw can be exploited by sending malformed multipart requests with arbitrary characters at the end of the boundary, leading to excessive resource consumption and a complete denial of service for all users. The vulnerability is unauthenticated, meaning no user login or interaction is required for an attacker to exploit this issue.

Multiple Server-Side Request Forgery (SSRF) vulnerabilities were identified in the significant-gravitas/autogpt repository, specifically in the GitHub Integration and Web Search blocks. These vulnerabilities affect version agpt-platform-beta-v0.1.1. The issues arise when block inputs are controlled by untrusted sources, leading to potential credential leakage, internal network scanning, and unauthorized access to internal services, APIs, or data stores. The affected blocks include GithubListPullRequestsBlock, GithubReadPullRequestBlock, GithubAssignPRReviewerBlock, GithubListPRReviewersBlock, GithubUnassignPRReviewerBlock, GithubCommentBlock, GithubMakeIssueBlock, GithubReadIssueBlock, GithubListIssuesBlock, GithubAddLabelBlock, GithubRemoveLabelBlock, GithubListBranchesBlock, and ExtractWebsiteContentBlock.