Technical impact
FastGPT is an AI Agent building platform. Prior to version 4.14.10.4, it suffered from a broken access control flaw (IDOR/BOLA). The API validates the user’s team token but fails to confirm that the requested application ID belongs to that team. This means any authenticated user can supply a foreign appId and execute another team’s private AI workflows, resulting in cross‑tenant data exposure and unauthorized execution. The weakness corresponds to CWE‑284 (Improper Authorization) and CWE‑639 (Authorization Bypass Through User‑Controlled Key).
Affected Systems
The affected product is FastGPT by labring. Versions before 4.14.10.4 are vulnerable. Any environment where users are authenticated and have API access is at risk.
Risk and Exploitability
The CVSS score of 5.3 indicates a moderate severity. EPSS is not available, and the vulnerability is not listed in CISA’s KEV catalog. An attacker would need to be authenticated within a team and target the appId parameter; no special privilege beyond regular authentication is required. Because the flaw enables cross‑tenant code execution and data leakage, the potential impact is significant for confidentiality and integrity, although the attack requires user authentication. The risk is therefore moderate to high for organizations that host multiple teams within FastGPT.
