Markdown Task Collection Exporter

How to use Markdown Task Collection Exporter

Bring tasks from a small collection of original Markdown notes into one reviewable artifact. A bounded CommonMark token parser identifies list-item first paragraphs, so examples inside fences and raw HTML blocks do not become ordinary tasks. Complete original files stay in the JSON evidence, while selected task labels retain their Markdown source rather than becoming executed markup.

1. Select one to ten distinct-name UTF-8 notes or load the original generated files.
2. Choose JSON source evidence or CSV task rows, then collect.
3. Review checked state, source paragraph lines and unassessed checkbox-like prefixes before sharing the separate export.

Good to know

• At most ten actual local files, 200,000 aggregate bytes and 50,000 decoded UTF-16 characters. File names at most 200 characters and distinct; UTF-8 is fatal-decoded with original BOM retained. Literal NUL is refused. No original file is edited.
• CommonMark block structure uses the pinned markdown-it 15.0.3 parser inside a same-origin module Worker. Preparation has a 10-second deadline; computation has a separate 2-second deadline. At most 10,000 aggregate tokens, 40 levels and 1,000 task/checkbox-like findings.
• Selected checkbox profile: the first paragraph of a list item begins with optional ASCII spaces then brackets containing one ASCII space, one tab, x or X, followed by ASCII space/tab/newline or end. Other checkbox-like prefixes are unassessed. This is not full GFM extension conformance, a due-date parser, a Markdown editor or a task-management account.
• JSON retains complete original text. One leading BOM is ignored for parsing; paragraph line-ending normalization follows CommonMark; reported start/end lines are one-based and inclusive, with original paragraph lines shown using LF separators. CSV formula protection can add apostrophes while original text remains in JSON. No HTML rendering, requests, execution or persistent history occurs.
• Processing is local. Output is limited to 1,000,000 text characters unless a stricter file limit is stated.