Docs / source reviewed
jev-tool-permissions
Choose permission paths for AI SDK tool calls
How this project uses Jev
- Input
- Task, tool definitions, and pending calls
- Jev decides
- Automatic approval, human confirmation, blocking, or relevance
- Code executes
- Apply deterministic rules first, then enforce decision-based gates
Evidence and limitations
Model failures fall back to ask-human. Applications still need to define and validate their permission policies.
This project has not been run independently here. Author-reported results are not independently verified results.