React 19: detect & patch unnecessary <input> DOM writes

Executive Summary

Unnecessary writes to <input> elements—repeatedly setting value, selection, or attributes on every render—are a common source of input jank and wasted CPU cycles in large React apps, and these patterns become more visible and brittle around major framework shifts like React 19. Front-end engineers, performance teams, and SREs at mid-to-large product organizations face this problem routinely: a single inefficient component pattern can affect thousands of inputs and translate into measurable regressions in RUM traces and user complaints. You could build a focused developer tool that combines static analysis (lint rules, AST transforms and CI gates), RUM-correlated detectors that surface only user-impacting DOM writes, and AI-assisted codemods that open safe, minimal PRs with estimated latency/CPU impact and rollback hooks. The product would ship an IDE/DevServer plugin for immediate feedback, a CI step to block regressions, and runtime instrumentation to prioritize fixes across monorepos; enterprise add-ons could include bulk automated patching, audit logs, and vendor-safe whitelists. This is timely: the addressable market is roughly $6.0B (25M developers × $240/year) for developer tooling and performance tools, React 19-driven migration cycles create short windows of urgency, and broader RUM/observability adoption makes it easier to detect and justify paid remediation. Additionally, advances in model-driven codemods lower the cost of producing safe diffs, improving the revenue calculus (revenue potential score: 88/100). To stand out you must combine evidence-based prioritization (static + RUM), conservative auto-fixes with clear safety nets, and an open-source core to build trust, while integrating deeply into PR/CI workflows; the key challenges are maintaining very low false-positive rates (target <1–2%), handling diverse code styles and browser edge cases, and earning developer trust to accept automated PRs.