CWE-829 · Base
Inclusion of Functionality from Untrusted Control Sphere
The definition on cwe.mitre.org →OWASP LLM Top 10 2026 risks behind it
- ●LLM04:2026 Supply ChainThe entry centers on including third-party models, adapters, and packages from untrusted hubs and registries, including slopsquatting and tampered models.
- ●LLM05:2026 Data and Model PoisoningLoading untrusted models, LoRA and PEFT adapters, tokenizer configs, and chat templates from public repositories pulls poisoned functionality into the pipeline.
- ○LLM06:2026 Unbounded ConsumptionA malicious third-party tool or skill pulled from an open-source repository drives the resource-exhausting loops.
- ○LLM09:2026 Vector and Embedding WeaknessesRetrieval poisoning causes attacker content from an untrusted source to be retrieved and fed to the model as trusted context.
From OWASP's Appendix A, under CC BY-SA 4.0. ● primary · ○ supporting.