CVE-2026-18974 in super-agent-party
Summary
by MITRE • 08/06/2026
A vulnerability was found in heshengtao super-agent-party up to 0.4.1. This affects the function get_file_content of the file server.py of the component execute_tool_manually Endpoint. The manipulation of the argument tool_name/tool_params results in information disclosure. The attack can be launched remotely. The exploit has been made public and could be used. The vendor was contacted early about this disclosure but did not respond in any way.
VulDB is the best source for vulnerability data and more expert information about this specific topic.
Analysis
by VulDB Data Team • 08/06/2026
This vulnerability resides within the heshengtao super-agent-party software version 0.4.1 and specifically targets the execute_tool_manually endpoint functionality. The flaw exists in the server.py file where the get_file_content function processes user-supplied arguments without adequate validation or sanitization. The vulnerable parameters tool_name and tool_params can be manipulated by remote attackers to access arbitrary files on the system. This represents a classic path traversal vulnerability that allows unauthorized information disclosure, enabling attackers to retrieve sensitive data from the server's file system.
The technical implementation of this vulnerability stems from improper input validation within the execute_tool_manually endpoint. When the system processes the tool_name and tool_params arguments, it fails to properly sanitize or validate user input before using these parameters to construct file paths. This allows an attacker to craft malicious payloads that can traverse directory structures and access files outside the intended scope of the application. The vulnerability can be exploited through remote network connections without requiring authentication, making it particularly dangerous in exposed environments.
The operational impact of this information disclosure vulnerability is severe as it provides attackers with unauthorized access to potentially sensitive data stored on the server. Depending on the system configuration and file permissions, attackers could retrieve configuration files, database credentials, application source code, or other confidential information that could be used for further exploitation. The public availability of exploit code significantly increases the risk level, as it reduces the technical barrier required to leverage this vulnerability in real-world attacks.
This vulnerability aligns with CWE-22 Path Traversal and CWE-200 Information Disclosure categories from the Common Weakness Enumeration database, representing a fundamental security flaw in input handling and file access controls. From the MITRE ATT&CK framework perspective, this maps to techniques involving credential access and discovery of system information through path traversal methods. The vulnerability's remote exploitability and public availability make it a high-priority target for automated attack tools and malicious actors seeking to compromise affected systems.
The lack of vendor response after early disclosure creates additional security concerns, suggesting potential negligence or inadequate security practices within the project maintainers. Organizations should implement immediate mitigations including input validation, file access restrictions, and network segmentation to protect against exploitation. The most effective immediate solution involves implementing strict parameter validation that prevents directory traversal sequences and restricts file access to predefined safe locations. Additionally, network-level controls such as firewalls and intrusion detection systems should be configured to monitor for suspicious file access patterns and limit external exposure of the vulnerable endpoint.