Packages

SuperOps REST API for modern managed service provider (MSP) platform combining RMM and PSA. Enables AI agents to manage IT support tickets and SLAs, access device inventory and health monitoring, manage client accounts and projects, handle invoicing and billing, trigger remote monitoring and scripting, integrate with documentation tools, and connect MSP operations with external business systems.

Next-generation browser and mobile test automation framework built on WebDriver and WebDriver BiDi protocols. WebdriverIO v8 supports both traditional WebDriver (Selenium Grid) and Chrome DevTools Protocol (CDP). Includes WDIO test runner, automatic browser management, Mocha/Jasmine/Cucumber integration, and native mobile support via Appium. More batteries-included than raw Selenium.

Database migration library and CLI for Go supporting 20+ databases including PostgreSQL, MySQL, SQLite, MongoDB, CockroachDB, and cloud databases. Migrations are versioned SQL or Go files (up/down). Available as both a CLI tool and Go library for embedding in applications. Widely used as the standard migration tool in Go microservices.

JavaScript import maps for Rails — enables ES module imports without a JavaScript bundler (no webpack, esbuild, or rollup). importmap-rails features: config/importmap.rb for pin declarations (pin 'stimulus', to: 'stimulus.min.js'), CDN pinning (pin_all_from, to CDN URLs), local JavaScript file serving via Sprockets/Propshaft, importmap:pin generator for adding packages from JSPM CDN, Rails 7+ default for new apps, Stimulus and Turbo integration, and asset cache busting via content-hash fingerprinting. Eliminates Node.js and npm as build requirements for agent web interfaces.

Pure Python RabbitMQ client library — AMQP 0-9-1 implementation for connecting to RabbitMQ. pika features: BlockingConnection for synchronous use, SelectConnection for async event-driven use, AsyncioConnection for asyncio, channel.basic_publish() for sending messages, channel.basic_consume() with callback for receiving, channel.queue_declare() with durable=True, channel.exchange_declare() for fanout/direct/topic exchanges, channel.queue_bind() for routing, basic_ack/nack/reject for message acknowledgment, prefetch_count for QoS, heartbeat_interval, SSL/TLS support, and connection parameters with credentials.

Modern, memory-safe TLS implementation in pure Rust. Supports TLS 1.2 and TLS 1.3. Does not use OpenSSL — eliminates OpenSSL's C code vulnerabilities. Used by major projects including the Rust toolchain, wasmtime, and the Let's Encrypt certbot replacement. Integrates with tokio-rustls for async TLS and reqwest for HTTPS clients.

Client-side collection agent that gathers host metrics, logs, traces, and process data and ships them to the Datadog SaaS platform for unified observability.

Descope — no-code/low-code authentication and user management platform with drag-and-drop flow builder for designing auth journeys (MFA, passkeys, SSO) plus REST API and SDKs for integration.

Make (formerly Integromat) REST API for managing scenarios (workflows), executions, and teams — enabling programmatic control of visual workflow automation across 1000+ integrated apps.

Mongoose is a MongoDB Object Document Mapper (ODM) for Node.js that provides schema-based modeling, built-in type casting, query building, middleware hooks, and validation on top of the native MongoDB driver.

Cross-platform ML model inference engine — runs ONNX format models on CPU, GPU, or specialized hardware with graph optimizations. ONNX Runtime features: InferenceSession (load .onnx model), session.run() for inference, execution providers (CPUExecutionProvider, CUDAExecutionProvider, TensorRTExecutionProvider, CoreMLExecutionProvider), graph-level optimizations (constant folding, operator fusion), quantization (INT8/FP16), model profiling, dynamic shapes, IoBinding for zero-copy GPU inference, and Python/C++/C#/Java/JavaScript APIs. Exports from PyTorch (torch.onnx.export), TensorFlow, and scikit-learn via skl2onnx. Microsoft production inference runtime — used by Azure ML, Windows ML, and Hugging Face Optimum.

Linear and Integer Programming modeling library for Python — formulate and solve LP/MIP problems with a simple Python API. PuLP features: LpProblem() for problem creation, LpVariable() for decision variables (continuous, integer, binary), lpSum() for objective/constraint expressions, status checking (LpStatus), solver integration (CBC default, GLPK, CPLEX, Gurobi, SCIP), LpMinimize/LpMaximize sense, variable bounds, constraint naming, results via .varValue, and JSON/MPS/LP file export. More accessible than OR-Tools for pure LP/MIP problems. Used for scheduling, routing, cutting stock, and resource allocation optimization.

Pydantic Logfire is an OpenTelemetry-based observability platform by the Pydantic team with first-class Python integration. It provides a simple Python SDK with automatic instrumentation for FastAPI, SQLAlchemy, HTTPX, OpenAI SDK, and other Python libraries, exporting traces and logs to the Logfire cloud dashboard.

Official JavaScript/TypeScript client library for Supabase — the open-source Firebase alternative built on PostgreSQL. Provides a unified client for Supabase's services: PostgreSQL database queries (via PostgREST), Auth (sign-in/JWT/OAuth), Storage (file upload/download), Realtime (websocket subscriptions to DB changes), and Edge Functions. Works in browser, Node.js, React Native, and edge runtimes. TypeScript-first with auto-generated types from your DB schema.

Python dataclass deserialization library — converts dicts/JSON into typed Python dataclasses with validation. dacite features: from_dict(DataClass, data_dict) converts nested dicts to dataclasses, strict=True rejects unknown keys, type_hooks for custom type conversion ({str: str.strip}), forward references support, Union type handling, Optional detection, nested dataclass support, and config object for transformation rules. Fills the gap between Python dataclasses (no built-in deserialization) and heavier ORMs/validators like Pydantic — useful for agent code parsing structured LLM responses or API payloads into typed dataclasses.

Python test data factory library for Pydantic, dataclasses, and attrs models — generates realistic mock data based on type annotations. polyfactory features: ModelFactory for Pydantic BaseModel, DataclassFactory for dataclasses, AttrsFactory for attrs classes, TypedDictFactory, build() for single instance, batch() for multiple, __faker__ for locale-aware data, __random_seed__ for reproducible data, field overrides in build() call, __set_as_default_factory_for_type__ for global registration, and constrained types (constr, conint) respected. Successor to pydantic-factories. Generates valid test data from type annotations without manual fixture construction.

PyTorch audio processing library — audio I/O, signal transforms, and pretrained models for speech and audio ML. torchaudio features: torchaudio.load() and save() for audio files (WAV, FLAC, MP3, OGG), torchaudio.transforms (MelSpectrogram, MFCC, Spectrogram, Resample, AmplitudeToDB, FrequencyMasking, TimeMasking), torchaudio.functional for signal processing ops, torchaudio.datasets (LibriSpeech, SPEECHCOMMANDS, VoxCeleb), pretrained models (Wav2Vec2, HuBERT via torchaudio.pipelines), StreamReader for streaming audio, and GPU-accelerated transforms. PyTorch audio companion — pairs with Whisper and HuggingFace speech models for agent audio pipelines.

Distributed stateful stream processing framework for real-time data pipelines with exactly-once semantics and event-time processing.

GraphQL client for JavaScript/React — handles queries, mutations, subscriptions, caching, and state management for GraphQL APIs. Apollo Client 3 features: useQuery() and useMutation() React hooks, InMemoryCache with automatic normalization, cache-first / network-only / cache-and-network fetch policies, optimistic updates (optimisticResponse), Apollo Link chain (auth, error, retry), subscription via WebSocket (subscribeToMore), refetchQueries after mutations, field-level cache policies, reactive variables (makeVar()), pagination helpers (fetchMore, relayStylePagination), and Apollo Client DevTools. Standard GraphQL client for React agent frontends querying agent GraphQL APIs.

Aviationstack real-time flight data REST API for travel apps, logistics platforms, and aviation services to access live flight tracking, historical flight records, airport information, airline data, and global aviation status — covering 13,000+ airlines and 10,000+ airports worldwide with real-time departure/arrival data, flight status, and route information. Enables AI agents to manage real-time flight status for travel notification automation, handle flight tracking for logistics coordination automation, access airport data for facility information automation, retrieve airline information for carrier details automation, manage historical flights for analytics automation, handle departure/arrival data for schedule management automation, access flight routes for itinerary planning automation, retrieve aircraft data for fleet information automation, manage flight delay for disruption management automation, and integrate Aviationstack with travel apps, trip planners, and corporate travel platforms for end-to-end aviation data automation.