NETROOM
Excalidraw is a virtual whiteboard for diagrams in a hand-drawn style: architecture sketches, user flows, wireframes and anything else that is quicker to draw than to describe. It runs in the browser with no sign-up - open a tab and start drawing. There are libraries of ready-made shapes, export to PNG and SVG, collaboration over a link with end-to-end encrypted rooms, and files you can keep locally or in your own storage. Open source, with a self-hosted option for teams.
Squoosh compresses images in the browser: drop a file in, pick a format and a quality, download the result. A slider puts the original and the compressed version side by side, so you can see exactly where artefacts start and stop just before them. It handles a range of formats, each with its own controls - from a single quality dial to the codec's own parameters. Nothing is uploaded: the work happens on your device, and it runs offline once installed as an app. Built by Google Chrome Labs, open source, no sign-up.
CyberChef is a workbench for data that needs turning into something else: decode Base64, hash it, parse an X.509 certificate, unpack an archive, undo an XOR, convert an encoding. Operations stack into a recipe - a chain the data runs through, recalculated as you adjust the parameters. It earns its keep when you hold an unidentified blob and need to work out what it is. Everything runs in the browser, nothing reaches a server, and you can download it to keep. Built at GCHQ, open source.
Stirling PDF is fifty-odd PDF operations in one place: merge and split, rotate pages, compress, sign, redact, run OCR, convert to other formats and back. It works in the browser, but the files stay with you: the service runs on your own machine, one Docker command away. That is the whole point - not sending contracts and passports to someone else's server to rotate a page. There is an API for wiring the processing into your own pipelines. Open source.
Weather by coordinates, city, or postal code: current conditions, hourly and 16-day forecasts, historical observations, and climate normals. Separate endpoints cover air quality, lightning strikes, severe weather alerts, and agriculture metrics. Sign up for a key - there's a permanently free tier for hobby use plus a business trial capped at 1,500 requests a day. Production traffic means a paid plan.
Japanese city weather from livedoor: current conditions, today's and tomorrow's outlook and temperatures, no key required, JSON responses. The service is gone: weather.livedoor.com/weather_hacks/ now redirects to the livedoor homepage and the API no longer exists. An independent fork at weather.tsukumijima.net mirrors the old response format, but it's a third-party effort with no guarantees. For anything new, take official JMA data or another provider.
Marine and weather data by coordinates: waves, swell, tides, wind, pressure, water and air temperature. The distinguishing trick is that one response carries values from several forecast models at once - NOAA, ECMWF, DWD, SMHI and others - so you can compare them side by side in a single JSON payload. Beyond forecasts there is history, tide extremes and sea-bio parameters. A key comes with registration; the free tier caps daily requests, which is fine for a prototype but not for production traffic.
Marine weather by coordinates: wave height, period and direction, wind, water and air temperature, currents, tides and sea level, both forecast and archive. The distinctive part is that each value arrives from several forecast models at once, leaving you to pick a source or average them. A key comes with registration, and the free tier allows only a small number of daily requests - fine for a side project, paid plans for anything real.
Weather by coordinates, city name or postal code: current conditions, hourly and multi-day forecasts, historical data, air quality, geocoding and raster map tiles for precipitation and temperature. JSON responses with client libraries for nearly every language. The free key activates within about half an hour of signup and covers the basic endpoints under a daily cap; deep history and minute-by-minute forecasts sit behind paid plans. The default choice for a weather widget or a course project.
UV index by latitude and longitude: current and clear-sky UV, the daily maximum with the time it hits, sunrise, sunset and twilight times, safe-exposure estimates per skin type, plus forecast and historical readings. JSON responses, with the key passed in a header and handed out immediately on signup. The free tier runs on a small daily quota - plenty for a sun-safety or gardening app - and heavier use moves to paid plans. Coverage is global, since it is computed rather than station-based.
Weather forecasts by coordinates with no signup at all: hourly and daily values - temperature, precipitation, wind, cloud cover, pressure, humidity, UV index - up to two weeks ahead. You can pick a specific numerical model or take a blended forecast, and sibling endpoints cover the historical archive back to 1940, air quality and marine conditions. A key is only needed for commercial use; non-commercial access is open. Responses are JSON and the variable set is chosen right in the query string.
Climate records from the US weather service: daily and monthly station observations, temperature, precipitation, snow, normals and extremes, filtered by dataset, data type, location or station. The endpoints are /datasets, /datatypes, /locations, /stations and /data. A free token is required and travels in a header; limits are five requests per second and ten thousand per day, with at most a thousand records per response. The service moved from ncdc.noaa.gov to ncei.noaa.gov, so double-check the host.
Official forecasts and alerts from the US National Weather Service: coordinates resolve to a forecast grid, from which you get period and hourly forecasts, observations from nearby stations, active storm and flood alerts, and radar station metadata. No key required, but a User-Agent with contact details is mandatory. Responses are GeoJSON and coverage is US and territories only. The data is public domain, so production use carries no strings.
Forecast and climate data from the Norwegian Meteorological Institute by coordinates: hourly forecasts nine days out, precipitation nowcasts, sunrise and sunset times, ocean state and severe-weather alerts. No key required, but you must send a meaningful User-Agent with contact details - anonymous traffic gets blocked. Rate limits are lenient provided you cache and honour Expires headers. Openly licensed data, fine for hobby work and production alike, with attribution.
High-resolution weather forecasts: over a hundred variables - temperature, precipitation, wind at multiple altitudes, cloud cover, solar radiation, plus indices tuned for agriculture and energy. The window runs fourteen days ahead and four days back for any coordinate pair. Keys come with registration, but this is a commercial product: expect a trial, then volume-based pricing. Aimed at production work where forecast accuracy matters more than cost per call.
Weather data from Germany's national meteorological service: observations and forecasts, readings from individual stations, and warnings issued separately for regions, the coast, the sea and the Alps. No key or registration, with an OpenAPI description available - though some data arrives as files and archives rather than tidy JSON. Coverage is mainly Germany and Europe. An official source with no commercial limits, usable in production once you have worked through the formats.
Weather by city name, coordinates, IP or postcode: current conditions, forecast, history and astronomy data such as sunrise and sunset. One thing to know first: APIXU was retired and the domain and docs now redirect to weatherstack, which is where keys and endpoints come from. A key is issued at signup; the free plan is capped monthly and excludes the forecast, which sits behind paid tiers. Old integrations pointing at apixu.com will not work.
Numerical weather forecasts by coordinates, including an astronomy-focused one: cloud cover, atmospheric transparency, seeing, humidity, wind and temperature - the set you actually need to plan a night at the telescope. Derived from NOAA and NCEP models. No key, no signup: a plain GET with latitude, longitude and a product type (ASTRO, CIVIL, METEO). Responses come as JSON or as a ready-made forecast image.
YouTube from code: search videos and channels, read video metadata, manage playlists, subscriptions, and comments, upload videos, and pull channel statistics through the Data API, with a separate Analytics API for reports. A key covers public reads; OAuth is needed to act on a user's behalf. Everything runs against a 10,000-unit daily quota - a search costs 100 units, a plain read costs 1 - so budget the calls up front.
Full access to the video platform: uploads, metadata and thumbnails, playlists and channels, privacy settings, player embedding, view statistics, and live streaming on higher tiers. Auth is OAuth 2.0 - register an app in the developer portal, though a personal access token is enough for your own videos without an interactive login. Official libraries cover PHP, Python and Node. Rate limits follow your account plan, and several capabilities are paid-tier only.
A TV series database: show descriptions, seasons and episodes with air dates, cast and characters, networks and streaming services, artwork, the schedule for a given day, and search by title or person. No key is needed - all read endpoints are open and return JSON, which makes it an easy stand-in for paid film databases in learning projects. A soft limit of roughly 20 calls per 10 seconds applies; commercial use is paid.
Movie and TV data plus a user's watch history: a catalogue of films, shows, seasons and episodes with ratings and synopses, trending and popular lists, an episode release calendar, and - once authorised - check-ins, custom lists, collection and watched state. Public data needs only a client id in a header; acting on a user's behalf requires OAuth. Rate limits apply and the JSON responses are well documented. A good fit for a show tracker when you would rather not store viewing progress yourself.
Movies and TV in depth: synopses, ratings, release dates, genres, cast and crew, trailers, posters and backdrops in multiple sizes, seasons and episodes, popularity and trending lists. The data is community-maintained with translations in dozens of languages. A key is free after registration and the terms are generous for non-commercial work - no hard daily ceiling, just a sane rate limit. It is effectively the default choice for media libraries and film apps, with client libraries in every major language.
A reference for the Star Trek universe: characters, performers, episodes, seasons, series and movies, spacecraft and their classes, planets, species, technology and organisations. Single-entity endpoints sit alongside paginated search, responses are JSON, and an OpenAPI spec plus generated clients are provided. No key, no signup. The dataset is large and carefully modelled - a rare fan API solid enough to build a real filterable catalogue on.
South Park character quotes over plain REST: a random line, or a selection filtered by character, returned as JSON. No key, no signup, and few enough endpoints to understand in five minutes. It is a small hobby project kept alive by one maintainer, with open sources you can redeploy yourself. Good filler for a learning frontend, a bot or a fetch-on-click demo; do not build production on someone else's hobby uptime.
Quotes and film data from the Shrek franchise: list the movies, pull a random quote, get every line from one film, or fetch top cast members. Plain REST with JSON responses, no key and no signup. The fourth entry exposes noticeably fewer endpoints than the rest, so plan around the first three. A fine data source for a learning frontend, a fetch demo or a random-quote bot; cache responses locally if anything depends on it.
Cloud video editing over an API: you describe a timeline in JSON - clips, titles, transitions, audio, templates - and the service renders the file and hands back a URL. Rendering is asynchronous, with status by polling or webhook, and separate APIs cover asset generation and template management. A key comes with signup, a sandbox environment is provided for testing, and production renders bill by output minutes. Good fit for data-driven video.
An Arabic streaming guide: 14,000+ movies and series with Arabic and English titles, year, genres, rating, poster path and which of 18 regional platforms carry them. The endpoints are static files - /api/titles.json, /api/platforms.json, /api/stats.json and per-letter search. No key, no signup, and no rate limits since nothing is computed per request. Free for personal and commercial use as long as you credit the source.
Editors' picks