JSON Repair Tool — Fix Broken & Malformed JSON Online

Popular Text & Data tools

• jq Playground & Filter BuilderInteractive jq expression tester, filter builder, and JSON transformer powered by WebAssembly.
• Base64 Encode/DecodeEncode and decode Base64 strings, files, and data URIs instantly.
• Git Conflict ResolverVisually resolve Git merge conflicts by picking current, incoming, or both versions.
• Text Case ConverterConvert text between 20+ case formats: camelCase, snake_case, kebab-case, Title Case, and more.
• Markdown PreviewPreview and render Markdown with GFM, math, Mermaid diagrams, and export options.

How to JSON Repair Tool Online

1. 1

   Paste your malformed or broken JSON into the input editor on the left — whether it originated from an LLM prompt (ChatGPT, Claude, DeepSeek), an API response, a server log, or a config file.

2. 2

   Observe the real-time syntax detector highlight structural issues: trailing commas, single quotes, unquoted keys, markdown code fences (```json ... ```), missing commas, or unescaped control characters.

3. 3

   Click the "Repair" button or press ⌘↵ (Ctrl+Enter on Windows/Linux) to trigger the automated 10-pass AST repair pipeline.

4. 4

   Review the repaired, RFC 8259-compliant JSON output in the right panel with full syntax highlighting and structured indentation.

5. 5

   Inspect the "Repair Log" audit tray below the editor to examine each repair strategy applied, including exact line and column coordinates of modifications.

6. 6

   Toggle between 2-space, 4-space, or minified formatting to optimize the payload for debugging readability or minimal wire transfer size.

7. 7

   Switch to the "Code Generator" tab to export the exact repair logic as standalone, dependency-free functions in TypeScript, Python, Go, Rust, Java, C#, PHP, or Ruby.

8. 8

   Copy the clean JSON with ⌘⇧C, download the output as a `.json` file, or pipe it directly into downstream tools like JSON Schema Visualizer or JSON to TypeScript.

JSON Repair Tool Features

• ✓

  10-Pass Deterministic Repair Pipeline: Sequentially applies targeted regex and lexical scanning passes to recover valid JSON structures without data loss.

• ✓

  AI Markdown Fence Stripper: Automatically strips ` ```json ` and ` ``` ` code fences and conversational preamble text emitted by LLMs like ChatGPT, Claude, and Gemini.

• ✓

  Trailing Comma Eliminator: Removes forbidden trailing commas from arrays (`[1, 2,]`) and objects (`{"a": 1,}`) to satisfy strict RFC 8259 parsers.

• ✓

  Single-to-Double Quote Normalization: Converts single-quoted strings and keys (`{'id': '123'}`) to standard double quotes while preserving escaped inner quotes.

• ✓

  Bare Object Key Quoter: Detects unquoted JavaScript-style property names (`{ name: "DevFlow" }`) and safely wraps them in compliant double quotes.

• ✓

  Missing Comma Insertion: Heuristically detects adjacent properties and array items separated only by whitespace or newlines and inserts necessary comma delimiters.

• ✓

  Truncated JSON Auto-Closure: Calculates unbalanced brace (`{`) and bracket (`[`) counts from incomplete streaming completions and appends missing termination tokens.

• ✓

  Concatenated Object Wrapper: Transforms multiple adjacent root-level JSON objects (`{"a":1}{"b":2}`) into a valid unified JSON array (`[{"a":1},{"b":2}]`).

• ✓

  JavaScript Comment Stripper: Removes single-line (`//`) and multi-line (`/* */`) comments common in JSONC files and configuration presets.

• ✓

  Control Character Escaping: Automatically escapes raw newlines, carriage returns, and tab characters embedded within literal string values.

• ✓

  Multi-Language Code Generator: Emits ready-to-use, zero-dependency repair algorithms in TypeScript, Python, Go, Rust, Java, C#, PHP, and Ruby for programmatic API pipelines.

• ✓

  Interactive Change Audit Log: Provides line-by-line and column-level tracking for every modification made during the repair cycle for full transparency.

• ✓

  Configurable Indentation & Minification: Format repaired payloads with 2-space, 4-space, or compact zero-whitespace single-line serialization.

• ✓

  Keyboard-First Workflow: Supports rapid developer iteration with ⌘↵ (Repair), ⌘⇧K (Clear), and ⌘⇧C (Copy output) hotkeys.

• ✓

  100% In-Browser Privacy: All parsing, string transformations, and AST repair operations execute locally in WebAssembly and JavaScript — zero network payload transmission.

Supported Formats & Dialects

The JSON Repair Tool supports 6 syntax formats and dialects for accurate parsing and processing.

Standard RFC 8259 JSON

Strict IETF JSON format requiring double-quoted keys and string values, strict comma separation, no comments, and zero trailing commas.

AI & LLM Markdown Fenced Blocks

JSON code blocks emitted by LLMs (ChatGPT, Claude, Gemini, DeepSeek) wrapped in ```json ... ``` fences with conversational preamble text.

JavaScript Object Literals & JSON5

Loose ECMAScript object literal syntax featuring single quotes, unquoted bare identifiers, trailing commas, and unescaped strings.

JSON with Comments (JSONC & VS Code Configs)

JSON configurations containing single-line (//) and multi-line (/* */) comments common in VS Code settings and tsconfig.json files.

Concatenated & Streamed Objects (NDJSON / Event Streams)

Consecutive un-bracketed JSON root objects ({...}{...}) produced by streaming SSE endpoints, audit log aggregators, and chunked exports.

Truncated & Incomplete Payloads

Prematurely severed JSON documents missing trailing braces, brackets, or closing string quotes due to LLM token context exhaustion or network cuts.

Frequently Asked Questions

Why do LLMs (ChatGPT, Claude, Gemini) produce malformed JSON and how does JSON Repair fix them?

Large Language Models emit tokens probabilistically. Even when instructed to output raw JSON, models frequently wrap output in markdown code fences (```json ... ```), include conversational preambles ("Here is your requested JSON:"), insert JavaScript comments, or run out of generation tokens before closing open brackets and braces. JSON Repair uses a 10-pass deterministic normalization pipeline that extracts balanced JSON structures, strips markdown formatting, and synthesizes missing closing tokens to make outputs 100% parseable by standard JSON.parse() engines.

How does the repair engine detect and remove trailing commas in arrays and objects?

RFC 8259 explicitly prohibits trailing commas before closing braces (}) and brackets (]). When developers copy JavaScript object literals or merge JSON snippets, trailing commas frequently trigger syntax errors in standard backend parsers. JSON Repair scans the document outside of quoted string literals using contextual regex scanning, identifies comma tokens immediately followed by closing delimiters (with optional whitespace or newlines), and strips them while preserving valid interior commas.

How does JSON Repair safely convert single quotes to double quotes without corrupting inner apostrophes?

Converting single quotes to double quotes requires distinguishing between quote delimiters and inner English apostrophes or contraction marks (e.g. "don't"). JSON Repair uses an escape-aware token scanning routine that identifies single quotes enclosing key names and string values, normalizes internal escaped characters (converting \' to ' and unescaped " to \"), and wraps the literal in standard RFC 8259 double quotes.

How are unquoted / bare object property keys identified and quoted without breaking JavaScript reserved words?

In JavaScript and JSON5, developers often write bare identifiers like { id: 101, status: "active" }. Standard JSON requires all property keys to be double-quoted strings ("id", "status"). The repair engine detects word tokens followed by a colon delimiter (:) that precede property values, validating that the token does not reside inside an already-quoted string or numeric literal, and automatically wraps the key in double quotes.

How does the tool auto-close truncated JSON caused by LLM token limits or network interruptions?

When an LLM exhausts its `max_tokens` budget or a streaming HTTP connection drops, the emitted JSON is cut off mid-stream, leaving unclosed strings, objects, and arrays. JSON Repair tracks the stateful nesting depth of open curly braces ({) and square brackets ([) across the document. It closes any unterminated string literal, and then appends the exact sequence of closing brackets and braces needed to produce a syntactically valid JSON fragment.

What is the architectural difference between a JSON Formatter/Linter and a JSON Repair engine?

A standard JSON Formatter or Linter relies on native JSON.parse() to ingest input; if the input contains a single syntax error, the parser throws an immediate unrecoverable exception and halts. In contrast, JSON Repair implements an error-tolerant pre-parser and AST healing engine that diagnoses 10 common classes of syntax corruption, applies heuristic corrections to produce valid JSON, and only then passes the normalized payload to the formatter.

How does JSON Repair handle JavaScript comments (// and /* */) in JSONC configurations?

While configuration files like tsconfig.json and VS Code settings.json support JSONC (JSON with Comments), standard JSON parsers reject any comment syntax. JSON Repair uses a tokenizer that tracks string literal boundaries to safely strip single-line (//) and multi-line (/* */) comments without modifying comment-like characters that legitimately appear inside string values (such as URLs like "https://example.com").

How does the tool detect and convert concatenated root objects into valid JSON arrays?

Log pipelines, event streaming endpoints, and NDJSON dumps often aggregate multiple JSON objects sequentially without commas or enclosing array brackets (e.g. {"event":"a"}{"event":"b"}). Standard JSON parsers fail because JSON permits only a single root document. JSON Repair detects adjacent object closures followed by opening braces (}{), inserts comma separators, and encapsulates the entire collection within a top-level array ([...]).

How does multi-language code generation work for programmatic pipeline integration?

The "Code Generator" tab generates clean, standalone, zero-dependency source code replicating the repair pipeline in TypeScript, Python, Go, Rust, Java, C#, PHP, and Ruby. You can select specific repair strategies (such as markdown fence stripping and trailing comma removal) and copy the generated functions directly into your backend services or ETL pipelines to sanitize incoming AI payloads before calling native JSON deserializers.

How does client-side browser execution ensure zero data retention for sensitive API payloads and tokens?

All repair passes, AST parsing, regex transformations, and code generation routines execute strictly within your local browser memory using client-side JavaScript. No payload data, API keys, credentials, or proprietary documents are ever uploaded to a remote server, external database, or third-party telemetry service, ensuring strict compliance with enterprise privacy, HIPAA, and GDPR standards.

Related Developer Tools

• JSON FormatterPrettify, minify, and validate JSON data instantly.
• JSON to Pydantic ModelConvert JSON and JSON Schema into typed Pydantic v2 and v1 BaseModel classes for FastAPI, Django Ninja, and Python applications.
• JSON to C# ClassConvert JSON to C# classes with System.Text.Json or Newtonsoft serialization.
• JSON to Go StructConvert JSON to Go structs with json tags and idiomatic naming.
• JSON to Rust StructConvert JSON into idiomatic Rust structs with Serde Serialize and Deserialize derive macros.
• JSON to TypeScript & Schema GeneratorGenerate TypeScript interfaces, Zod schemas, and Valibot schemas from JSON.
• MCP Schema ValidatorValidate, format, test, and debug Model Context Protocol (MCP) server schemas and tool definitions.
• JSON to TypeScriptConvert JSON to TypeScript interfaces or type aliases instantly.
• JSON to Zod SchemaConvert JSON to Zod schema definitions with automatic TypeScript type inference.
• Text Diff CheckerCompare two text blocks and highlight exactly what changed.
• JSONL ConverterConvert between JSONL (JSON Lines / NDJSON) and standard JSON arrays with auto-detection.
• TOML ConverterConvert between TOML and JSON formats with syntax validation, key sorting, and auto-direction detection.
• YAML ConverterConvert between JSON and YAML with validation, formatting, and multi-document support.
• AI Token CounterCount tokens and estimate API costs for 100+ LLMs with exact BPE and provider tokenization.
• Stack Trace AnalyzerParse, format, and analyze error stack traces with root cause isolation.