Goal

Move browser/input parity out of the current RC blocker set and into post-RC follow-up work without lying about the broader blanket replacement claim.

Scope

• live verdict docs only
• Target A / Target B blocker logic
• no browser/input proof rows are deleted

Loaded Skills

• major-task
• task
• goal workflow
• learnings-researcher

Phases

• Load source-of-truth docs
• Rewrite live verdict/control docs
• Run consistency sweeps
• Summarize the new blocker read

Findings

• the live docs currently treat browser/input parity as a blocker for both Target A and Target B
• the unresolved browser/input slice is already mostly external evidence: Android keyboard features plus broader iOS Safari composition/focus
• the coherent interpretation of “defer to post-RC” is:
  • Target A can stop treating browser/input parity as a current blocker
  • Target B still cannot honestly claim blanket zero-regression until that follow-up work lands or the broader claim is explicitly narrowed

Progress

2026-04-15

• read the live verdict stack:
  • release-readiness-decision.md
  • master-roadmap.md
  • overview.md
  • true-slate-rc-proof-ledger.md
  • replacement-gates-scoreboard.md
• rewrote the live verdict stack so:
  • Target A returns to Go
  • browser/input parity becomes post-RC follow-up instead of a current RC blocker
  • Target B stays challenged by exhaustive API/public-surface audit and the still-unclosed broader browser/input claim
• ran consistency sweeps across the live verdict docs

Errors

• none yet