PasteKit

PasteKit is a private online formatter, validator and converter that runs entirely inside the browser. Paste JSON, JSON5/JSONC, YAML, XML, CSV, TSV, TOML, SQL, HTML, Markdown, CSS, JavaScript, TypeScript or any of the other formats it supports, and PasteKit detects what it is looking at and pretty-prints it on the right-hand side in milliseconds. It is aimed at developers, data analysts and anyone else who regularly has to read, clean up or convert structured text, and its main purpose is to provide a fast formatting workspace where the pasted data never has to leave the machine. The everyday problem PasteKit addresses is that most online formatters work by shipping your clipboard contents somewhere else. What people paste tends to be sensitive: API responses containing customer records, configuration files with tokens, internal hostnames, database dumps, log excerpts or production SQL. PasteKit is explicit that everything is parsed, formatted and converted in a Web Worker inside the browser, and that the published Content Security Policy blocks the page from sending input anywhere. Alongside privacy, it addresses the smaller frustrations of online formatters: errors that do not tell you where the problem is, large integers that silently lose precision, and multi-megabyte files that freeze the tab. PasteKit points every error to its line and column with a plain-English explanation, keeps big numbers exact, and runs formatting in a background thread so large files stay smooth. The core of the product is formatting coverage: PasteKit formats, validates and converts 58 formats and 48 conversion pairs. The popular formats include JSON with a lossless streaming parser and printer, JSON5 and JSONC via Prettier, json5 and jsonc-parser, YAML through the eemeli yaml library covering YAML 1.2 and 1.1, XML through a built-in tokenizer and printer where DTD entities are never expanded, CSV and TSV through Papa Parse, TOML through taplo (WASM) for layout with smol-toml for validation, SQL through sql-formatter, and HTML, Markdown, CSS, JavaScript and TypeScript through Prettier, with html-minifier-terser, csso and terser available for minification. A broader set covers JSON Lines/NDJSON, SCSS, Less, GraphQL, INI, Java .properties, .env (dotenv), Dockerfile, Nginx config, Apache config, Python, Go, Java, PHP, C/C++/Objective-C, Shell/Bash, Vue single-file components, Angular templates, Handlebars/Glimmer, Lua, Dart, Zig and SVG. PasteKit states that it formats with the same engines developers run locally — Prettier, Ruff, gofmt, clang-format, rustfmt and others compiled to WebAssembly. Beyond formatting, PasteKit bundles 16 JSON tools. JSON Diff compares two documents by structure and reports added, removed, changed and moved values side by side as JSON Patch. JSON Patch applies or generates RFC 6902 JSON Patch and RFC 7396 merge patches, with failing operations pointing to their line. The JSON Schema Validator validates JSON or YAML against JSON Schema draft-04 to 2020-12 or an OpenAPI file, and can generate a schema from data. JSON Repair fixes broken JSON from LLMs, logs and JavaScript — code fences, quotes, commas, comments and truncation — listing each fix. JSON Viewer explores large JSON as a collapsible tree with type counts, search, depth control and one-click paths, while JSON Table Viewer shows an array of objects as a sortable, filterable table with flattening and export to CSV, Excel or Markdown. JSON Path Finder returns JSONPath, jq, JavaScript, Python and JSON Pointer paths; JSON Size Analyzer identifies which paths and repeated keys make a payload big; JSON Token Counter counts GPT tokens across pasted, minified, pretty, YAML and CSV forms to find the cheapest shape to send to an LLM; JSON Sort sorts keys at every level; and JSON Escape, JSON Stringify and curated Sample Datasets round out the set. The 48 converters move data between shapes rather than just prettifying it. JSON can be converted to YAML, XML, CSV, TOML, JSON Lines, SQL, JSON Schema, flattened JSON, Markdown tables or typed code in TypeScript, Go, Python, Java, Kotlin, C#, Rust, Swift and PHP. YAML converts to JSON, TOML and .env; XML converts to JSON and YAML; CSV converts to JSON, Markdown tables, SQL, HTML tables, TSV, and CSV/TSV convert between each other. Markdown converts to HTML and PDF, HTML to Markdown and JSX, cURL commands to JavaScript fetch, Axios, Python requests, Go net/http and PHP, and Unix timestamps to ISO 8601 and back. PasteKit also includes decoders and explainers: a JWT decoder using jose (decode only, verification opt-in) and a JWT Encoder for building and signing tokens with HS, RS, PS, ES or EdDSA keys, Base64/Base64URL decoding with format auto-detect, URL and query string parsing via WHATWG URL and URLSearchParams, HTTP request/response header parsing, a unified diff viewer powered by diff2html, a cron expression explainer built on cronstrue and cron-parser, and a regular expression explainer using regexp-tree with railroad diagrams. Specialist formatters cover Rust, C#, Kotlin, Swift, Ruby, HCL/Terraform, Protocol Buffers, LaTeX, DAX, Power Query M, KQL, Cypher, MongoDB shell queries, log lines and stack traces. The overall approach is deliberately local-first. Parsing happens in a Web Worker on your machine, which is what keeps large files smooth and keeps the data from being sent anywhere; the published Content Security Policy enforces that claim and can be verified in the browser's network panel. Share links keep their content after the # in the URL, which browsers never send to a server. After the first visit PasteKit works offline as an installable web app. In the editor you choose indentation (two spaces, four spaces or tabs), beautify or minify, sort keys and wrap lines, switch the output between Code, Tree and Table views, edit the output directly, and invoke commands with a ⌘K command palette. The benefits follow from those choices. You get one workspace instead of a different site for every format, so a YAML config, a SQL query and a JSON payload can all be handled in the same tab. Sensitive content can be pasted without a second thought, because nothing is uploaded. Mistakes are easier to fix because errors are located by line and column and explained in plain English, and identifiers such as large ledger IDs survive a round trip without precision loss. The converters remove the manual step of reshaping data for code generation or spreadsheets, and the JSON tools turn one-off debugging chores — diffing two payloads, generating a schema, repairing output from a model — into a single paste. Typical use cases described on the site include formatting and validating API responses and configuration files, repairing malformed JSON produced by LLMs or logs, diffing two JSON documents by structure, validating JSON or YAML against a JSON Schema or OpenAPI file, generating TypeScript, Go, Python, Java, Kotlin, C#, Rust, Swift or PHP types from a JSON sample, turning JSON into a real .xlsx or reading .xlsx and CSV into JSON, converting CSV into SQL or Markdown tables, decoding and inspecting JWTs, reading Apache, Nginx, syslog or JSON logs alongside Java, Python, .NET and JavaScript stack traces, and checking how many GPT tokens a payload costs in each serialization form. PasteKit is aimed at developers, data analysts, DevOps and platform engineers and anyone who handles structured text daily. It is web-based and installable as an offline app. Per the site's FAQ, all 58 formatters, 48 converters and 16 JSON tools are free to use, with no sign-up and no usage limits beyond what the browser can hold in memory. In short, PasteKit's value proposition is simple: format anything, upload nothing. It combines broad format coverage, a substantial JSON toolset and a deep converter catalogue inside a browser-only workspace, so the convenience of an online formatter no longer requires handing over the data you paste.