Directory Traversal in modelscope/agentscope
Description
A directory traversal vulnerability exists in modelscope/agentscope version 0.0.4. An attacker can exploit this vulnerability to read any local JSON file by sending a crafted POST request to the /read-examples endpoint.
AI Insight
LLM-synthesized narrative grounded in this CVE's description and references.
A directory traversal in AgentScope 0.0.4 allows unauthenticated attackers to read arbitrary local JSON files via a crafted POST to /read-examples.
Vulnerability
Overview
CVE-2024-8524 describes a directory traversal vulnerability in AgentScope version 0.0.4, specifically within the /read-examples endpoint. The application fails to properly sanitize user-supplied file path input, enabling attackers to escape the intended examples directory [2][3].
Exploitation
An unauthenticated attacker can exploit this vulnerability by sending a crafted POST request to the /read-examples endpoint with a path parameter containing directory traversal sequences (e.g., ../). This allows the attacker to read any local JSON file on the server, regardless of its location [2][4].
Impact
Successful exploitation enables the attacker to read sensitive information stored in JSON files, such as configuration parameters, authentication tokens, or application data. This information leakage could be leveraged for further attacks against the system or its users [4].
Mitigation
The vulnerability affects AgentScope 0.0.4. Users are advised to update to a patched version or manually implement input validation to restrict file paths to the intended examples directory. The project maintainers have been notified, and a fix is available in later releases [1].
AI Insight generated on May 20, 2026. Synthesized from this CVE's description and the cited reference URLs; citations are validated against the source bundle.
Affected packages
Versions sourced from the GitHub Security Advisory.
| Package | Affected versions | Patched versions |
|---|---|---|
agentscopePyPI | <= 0.0.4 | — |
Affected products
3- Range: =0.0.4
- modelscope/modelscope/agentscopev5Range: unspecified
Patches
0No patches discovered yet.
Vulnerability mechanics
AI mechanics synthesis has not run for this CVE yet.
References
4News mentions
0No linked articles in our index yet.