Packages

Open-source feature flag management system with a REST API for creating, enabling, and targeting feature toggles, available as self-hosted or managed SaaS (Unleash Cloud).

Vite-native unit test framework with Jest-compatible API. Vitest reuses the Vite config and plugin system, providing instant test startup, native ESM support, and TypeScript without transpilation. Uses the same describe/it/expect API as Jest but with significantly faster execution due to Vite's HMR-based test runner. Recommended for projects already using Vite.

Full-stack Python web framework with batteries included — ORM, admin interface, auth, templating, forms, migrations, caching, sessions, and signals. Django features: Model-Template-View (MTV) pattern, ORM (Model.objects.filter/create/update/delete), automatic admin interface (admin.site.register), URL routing (urls.py patterns), Django REST Framework integration, Channels for WebSocket (ASGI), class-based and function-based views, form validation, CSRF protection, middleware stack, management commands, Django's queryset lazy evaluation and select_related/prefetch_related for N+1 prevention, and database migrations via makemigrations/migrate.

Ruby test data factory library — replaces static fixtures with dynamic, flexible test object creation. factory_bot defines factories with default values and lets tests override only what matters. Key features: build (in-memory, no database), create (persists to database), build_stubbed (fake object, no database queries), traits (reusable attribute sets), sequences (unique values), associations (nested factory objects), callbacks (after_build, after_create), and transient attributes. factory_bot_rails integrates with Rails for automatic factory loading. Used in RSpec, minitest, and plain Ruby test suites for generating agent test data.

Run multiple promise-returning functions with limited concurrency. p-limit creates a concurrency limiter that ensures only a specified number of async operations run simultaneously. Essential for agent systems that make many parallel API calls but need to respect rate limits — process 1000 items with max 10 concurrent requests.

Pytest plugin that provides a mocker fixture wrapping Python's unittest.mock — the same capabilities as mock.patch/MagicMock but as a pytest fixture instead of decorator/context manager. pytest-mock's mocker fixture: creates mocks that are automatically cleaned up after each test (no need for patch.stop()), provides mocker.patch(), mocker.MagicMock(), mocker.spy(), mocker.stub(), and mocker.AsyncMock(). Integrates with pytest's assertion rewriting for clearer assertion failure messages. Preferred over unittest.mock decorators in pytest-based test suites for cleaner test function signatures.

URL slug generation library. Converts strings to URL-safe slugs by replacing spaces with hyphens, removing special characters, transliterating Unicode characters to ASCII, and lowercasing. Supports 200+ Unicode character transliterations (German umlauts, French accents, Cyrillic, Arabic, etc.). Used for generating SEO-friendly URLs from titles, blog post slugs, and file name sanitization.

Derive macro for implementing std::error::Error for custom Rust error types. Write error enums with #[derive(Error)] and #[error('description')] attributes — thiserror generates the Error implementation, Display, and From conversions automatically. The standard approach for defining typed errors in Rust library crates, complementary to anyhow for application code.

Structured, contextual logging and distributed tracing framework for Rust. Instruments async code with spans (logical units of work) and events (log messages with metadata). Works with multiple subscribers/collectors — tracing-subscriber for local logging, opentelemetry-rust for distributed tracing. The standard observability approach for async Rust services (Axum, Tokio use it internally).

Exhaustive pattern matching for TypeScript. ts-pattern provides a match() function that handles all cases of discriminated unions, union types, and complex data structures with compile-time exhaustiveness checking. The TypeScript compiler guarantees all cases are handled — missing a case is a type error. Replaces verbose switch statements with readable, type-safe pattern matching like Rust's match or Haskell's case expressions.

Ultra-fast Rust-based Python package and project manager that replaces pip, pip-tools, virtualenv, and pyenv with a single tool running 10-100x faster.

JavaScript/TypeScript tool for writing shell scripts. zx provides the $`command` template tag to execute shell commands from JavaScript with proper error handling, output capture, and async support. Eliminates shell scripting pain points — handles quoting, error codes, parallel execution, and piping while staying in JavaScript/TypeScript. Built by Google.

AgentUp is a framework for creating portable, scalable, and secure AI agents that can run consistently across different environments.

MCP server for Android device automation using a Puppeteer-style API. Enables AI agents to control Android devices — tapping, swiping, typing, taking screenshots, and navigating apps — through Android Debug Bridge (ADB). Supports UI testing, app automation, and AI-driven Android interaction workflows.

Terminal string styling library for Node.js. Chalk provides a chainable API for adding colors, bold, italic, underline, and background colors to terminal output. Handles terminal color support detection automatically — no color output in non-TTY environments unless forced. The most popular terminal styling library for Node.js CLI tools. Chalk 5+ is ESM-only.

Plugin marketplace for Claude Code comprising six Roman-themed plugins: Praetorian (context guard), Historian (session memory), Oracle (tool discovery from 17 sources), Gladiator (failure learning), Vigil (file recovery checkpoints), and Orator (deterministic prompt optimization). Plugins use hook-based event system to trigger MCP tools at high-impact moments.

CodeRide eliminates context reset between coding sessions by providing MCP integration for persistent codebase context. Enables AI coding agents to maintain understanding of a codebase across sessions — preventing the 'context reset' where agents lose project understanding when a conversation restarts.

GitHub MCP Server is an MCP-compliant server built with TypeScript to extend AI agents with GitHub integration, providing tools for repository management, branch operations, file handling, issue tracking, and collaboration features.

Google Calendar Streamable MCP server enabling AI agents to interact with Google Calendar via the Google Calendar API — creating, reading, updating, and deleting calendar events, managing multiple calendars, checking availability and free/busy times, and integrating Google Calendar scheduling into agent-driven personal assistant and workflow automation applications.

Google Contacts MCP server enabling AI agents to interact with Google Contacts via the People API — searching and retrieving contacts, reading contact details (email, phone, address), creating and updating contacts, and integrating Google's contact management into agent-driven personal assistant, CRM, and communication workflow automation.