Browse / Improper Output Handling / CVE-2026-6110
MetaGPT has an eval injection in metagpt/strategy/tot.py
- Class
- Improper Output Handling (filed by a model that read the advisory)
- Technique
- Code and command execution
- Affected
- metagpt (PyPI)
- CVSS
- CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:L
- EPSS
- 0.4% chance of exploitation in 30 days
How it works, in short
Model-generated thought text is passed to eval in the Tree-of-Thought solver, so crafted model output runs as code. (Our summary, written by a model. The source text is below.)
Details
A vulnerability was identified in FoundationAgents MetaGPT up to 0.8.2. This affects the function generate_thoughts of the file metagpt/strategy/tot.py of the component Tree-of-Thought Solver. The manipulation leads to code injection. It is possible to initiate the attack remotely. The exploit is publicly available and might be used. The project was informed of the problem early through an issue report but has not responded yet.
References
- https://nvd.nist.gov/vuln/detail/CVE-2026-6110
- https://github.com/FoundationAgents/MetaGPT/issues/1933
- https://github.com/FoundationAgents/MetaGPT/pull/1946
- https://github.com/FoundationAgents/MetaGPT
- https://vuldb.com/submit/791761
- https://vuldb.com/vuln/356970
- https://vuldb.com/vuln/356970/cti
Source: GitHub Advisory Database, used under CC BY 4.0.