D4rt
D4rt runs Dart source at runtime in a permission-guarded sandbox. It is the dynamic-execution engine that lets Tom's AI scripting and flow layers generate and run code safely.
D4rt is a family of Dart interpreters that execute Dart source at runtime without compilation, inside a permission-guarded sandbox. It is the layer that lets `tom_brain` and `tom_flow` run AI-generated and scripted code, and lets agentic UI render in a Flutter client.
Two parallel interpreter lines share one bridge API. `tom_d4rt` is the analyzer-based reference implementation. The analyzer-free line — `tom_ast_model`, `tom_d4rt_ast`, `tom_ast_generator`, `tom_d4rt_exec` — parses on a build machine and interprets on-device, so the heavyweight analyzer never ships inside an app.
Around the interpreters sit code generators (`tom_d4rt_generator` bridges native APIs), Flutter bridges (`tom_d4rt_flutter`, `tom_d4rt_flutter_ast`) for over-the-air UI, and DCli-equipped REPLs (`tom_d4rt_dcli`, `tom_dcli_exec`). D4rt underpins Tom's creation flow and the agentic-app line.
The d4rt family
tom_d4rt Published
tom_d4rt_flutter In the works
tom_d4rt_test Published
tom_d4rt_dcli Published
tom_d4rt_flutter_test In the works
tom_ast_model Published
tom_ast_generator Published
tom_d4rt_ast Published
tom_d4rt_exec Published
tom_dcli_exec Published
tom_d4rt_flutter_ast In the works
tom_d4rt_flutter_ast_test In the works
License
# License See each folder/package for its specific license.