What problem does it solve?
Verifying that existing Slate editor proof surfaces (tests, typechecks, Playwright browser suites) pass repeatably requires disciplined execution, timing measurement, and failure logging, which is tedious and error-prone to do by hand.
Core Features & Use Cases
- Repeated Gate Execution: Runs existing gate commands (bun check, package tests, Playwright suites) in a measured loop with elapsed-seconds metrics.
- Pass/Fail Logging: Records statuses such as measure, keep, checks_failed, crash, and discard, plus duration metrics and dashboard state.
- Failure Routing: Detects repeated failure signatures and hands off to the right owner (slate-patch for correctness bugs, slate-plan for design issues, slate-ar-perf for optimization).
- Use Case: After modifying editor navigation behavior, run the full Playwright navigation and typing suites repeatedly to confirm the proof surface passes consistently and capture flake evidence.
Quick Start
Ask the agent to run the slate-ar-gate loop on the full editor behavior suite and report pass/fail status with timing metrics.