This roadmap captures the next major project priorities.
- Benchmark coverage expansion
- Run
terminal-benchandswe-benchcontinuously. - Add more benchmark suites over time for broader capability coverage.
- Documentation site maturity
- Continue improving docs information architecture, content depth, and discoverability.
- Evaluation framework
- Build a reusable eval framework for scenario definition, execution, scoring, and regression tracking.
- Claude Code capability alignment
- Close feature gaps and align more behaviors with Claude Code where appropriate.
- Playground support
- Build and iterate a product playground for interactive testing and demos.