To prevent agents from obeying malicious instructions hidden in external data, all text entering an agent's context must be ...
Note: jsrun is under development. Expect breaking changes between minor versions. One of the most compelling use cases for jsrun is building safe execution environments for AI agents. When LLMs ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback