Tells you which component broke hydration, what differed, and how to fix it — in dev, with zero production cost.

An implementation of Amazon's DynamoDB built on LevelDB

A react-native solution to check if your app runs under battery optimization on Android devices and makes it easy for the user to disable it.

Shared JS/TS dev-tooling config (tsconfig, eslint, prettier, commitlint, lint-staged, semantic-release) for jabrown93 repos

The one definition of what a logger is on the OA Drive platform - four methods, validated fail-fast, no console fallback

Flexible visual snapshot testing for Playwright, built for e-commerce.

gettext-like client-side Javascript Internationalization Library

utility script for detecting and (optionally) fixing differences between multiple DynamoDB tables (in the same or different regions)

Production gates for Astro sites — content-hashed build ids, freshness watchdogs, performance budgets, discovery wiring, and a no-third-party-asset tripwire.

Canonical JSON Schemas for the WasmAgent Agent Evidence Protocol (AEP) and compliance contracts. Single source of truth across the WasmAgent org.

thin MCP server over the public driver API

Verify that a bug reproduction is stable and produce an agent-ready evidence receipt.

Framework-agnostic SDK for [Kitenzo](https://apps.shopify.com/bundlebuilder). Provides an API client, bundle configuration state machine, types, and utilities for building custom bundle experiences on any JavaScript platform.

A measured prose style for documentation, with a checker calibrated on human technical writing.

One Standard Schema validation adapter for Wizzard flows — Zod, Valibot, ArkType, Effect and Yup.

Optimize the tone of Pi prompts without changing their content

Audit an agent harness against the harness-evaluation criteria, with machine-enforced evidence validation.

Official TypeScript SDK for the agentleFS REST API.

MCP server that shows what is consuming your agent's context window before the conversation starts — skill descriptions, memory files and their imports, and configured MCP servers — ranked, with duplicates called out.

Gig/creator algorithmic fairness: decision receipts a worker can audit — completeness (every decision about me, tamper-evident), rule-consistency (the same rulebook decided everyone, via manifest_hash), and committed factors (no post-hoc rationalization).