Prompt injection attacks on agentic AI coding editors like Cursor and GitHub Copilot reach up to 84% success in executing malicious commands by poisoning external development resources.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
Fine-tuning and prompting reduce some CWEs in AI-generated code but frequently introduce new weaknesses, with no strategy working reliably across models or languages.
citing papers explorer
-
"Your AI, My Shell": Demystifying Prompt Injection Attacks on Agentic AI Coding Editors
Prompt injection attacks on agentic AI coding editors like Cursor and GitHub Copilot reach up to 84% success in executing malicious commands by poisoning external development resources.
-
On Fixing Insecure AI-Generated Code through Model Fine-Tuning and Prompting Strategies
Fine-tuning and prompting reduce some CWEs in AI-generated code but frequently introduce new weaknesses, with no strategy working reliably across models or languages.