A browser-local HTML5 character-reference workbench with context-aware escaping, all WHATWG named entities, decimal/hex references, legacy-reference diagnostics, recursive decode, double-encoding detection, round-trip verification, batch files and a searchable entity inspector.
The encoder distinguishes HTML text from quoted attributes, so it does not blindly escape everything. Existing valid references can be preserved to prevent double encoding. Every encode result is decoded again locally and compared against the intended semantic text before Copy/Download is enabled.
Engine ready. Ctrl/Cmd + Enter converts.
Search a character, entity name or Unicode code point.
Process multiple text/HTML files with the current converter settings. Files stay in this browser; successful outputs are packed into a ZIP locally.
No files selected.
Prevents & from becoming & when the source already contains a valid entity.
Off follows browser-compatible HTML5 legacy behavior. On decodes only references ending in ;.
Use 2+ only when intentionally unpacking strings such as <. One pass is safest.
Downloaded results are UTF-8 text.
The inspector never renders decoded HTML; dynamic values are displayed as text to avoid script execution.
Encoding, decoding, file reading, inspection and ZIP creation are local. This package contains no conversion endpoint, analytics call or account requirement.
Minimal safe escapes only characters required for the chosen HTML context. ASCII-safe additionally converts non-ASCII text. Named/decimal/hex profiles convert unsafe + non-ASCII characters using the selected notation.
Text: Tom & Jerry <3 Attribute: title="Tom & Jerry"
The decoder understands the complete HTML5 named-reference table plus decimal and hexadecimal numeric references, including the browser compatibility replacement rules for legacy numeric references.
© → © © → © 😀 → 😀
HTML parsing rules differ between text and attributes, especially for legacy named references without semicolons. Choose the destination context before encoding or auditing.
If text already contains &, encoding its ampersand again produces &amp;. Preserve Existing References is enabled by default, and the audit also detects likely double-encoded references.
For encoding, OneFreePlace immediately decodes the result with HTML5 rules and compares the semantic text. If they differ, output use is blocked and the audit explains why.
This tool works on character references, not full HTML sanitization. Encoding text does not make arbitrary HTML, URLs, CSS or JavaScript safe. Use a context-specific sanitizer when inserting untrusted markup.