CVE-2026-7812
Published: 05 May 2026
Summary
CVE-2026-7812 is a medium-severity Injection (CWE-74) vulnerability. Its CVSS base score is 5.5 (Medium).
Operationally, exploitation aligns with the MITRE ATT&CK technique Exploit Public-Facing Application (T1190); ranked in the top 15.5% of CVEs by exploit likelihood; it is not currently listed in the CISA KEV catalog.
This vulnerability is AI-related — categorised as AI Agent Protocols and Integrations; in the Protocol-Specific Risks risk domain.
The strongest mitigations our analysis identified are NIST 800-53 SI-10 (Information Input Validation) and SI-2 (Flaw Remediation).
Deeper analysis
A command injection vulnerability affects the git_operation function in the MCP Tool component of 54yyyu code-mcp up to commit 4cfc4643541a110c906d93635b391bf7e357f4a8. The flaw resides in src/code_mcp/server.py and stems from unsanitized handling of the operation argument, enabling arbitrary command execution. The product follows a continuous delivery model with rolling releases, so no specific affected or fixed versions are identified. The issue was reported via a public GitHub issue but has received no response from maintainers.
Remote attackers can exploit the vulnerability without authentication or user interaction by supplying a crafted operation argument, resulting in command injection with limited impacts on confidentiality, integrity, and availability. Public exploit code is available, and the weakness maps to CWE-74 and CWE-77.
The referenced GitHub repository and Vuldb entries provide no mitigation guidance or patches. The associated EPSS score has remained essentially flat between 0.0212 and a peak of 0.0218.
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2026-27217
Vulnerability details
A vulnerability was found in 54yyyu code-mcp up to 4cfc4643541a110c906d93635b391bf7e357f4a8. The impacted element is the function git_operation of the file src/code_mcp/server.py of the component MCP Tool. Performing a manipulation of the argument operation results in command injection. The attack can…
more
be initiated remotely. The exploit has been made public and could be used. Continious delivery with rolling releases is used by this product. Therefore, no version details of affected nor updated releases are available. The project was informed of the problem early through an issue report but has not responded yet.
- CWE(s)
AI Security AnalysisAI
- AI Category
- AI Agent Protocols and Integrations
- Risk Domain
- Protocol-Specific Risks
- OWASP Top 10 for LLMs 2025
- None mapped
- Classification Reason
- Matched keywords: mcp
Related Threats
MITRE ATT&CK Enterprise TechniquesAI
Why these techniques?
Command injection in a remotely accessible, unauthenticated MCP server component directly enables T1190 (Exploit Public-Facing Application) for initial access and results in arbitrary command execution that maps to T1059 (Command and Scripting Interpreter).
CVEs Like This One
Affected Assets
Mitigating Controls
Mitigating Controls (NIST 800-53 r5) AI
Directly prevents command injection by validating the user-controlled 'operation' argument in git_operation before execution to block malicious payloads.
Requires timely remediation of the known command injection flaw in code-mcp's git_operation function through patching or compensatory controls.
Limits the impact of injected commands by enforcing least privilege on the server process handling the vulnerable git_operation function.