Markdown Link Flattener
Turn `[label](url)` links into plain labels, `label (url)` citations, or URL-only lists. Images, reference definitions and code blocks have explicit controls, and everything runs locally in your browser.
About this tool
Markdown links are great while you are editing, but they get in the way when the text is headed to
plain email, a CMS field, a spreadsheet, an LLM prompt, or a legal/redline review that should show
the visible words instead of link syntax. This tool flattens inline Markdown links in place:
[the docs](https://example.com/docs) can become the docs, the docs (https://example.com/docs),
or just https://example.com/docs.
The surrounding text stays in the same order. Code spans and fenced code blocks are preserved by
default, so README examples that intentionally show Markdown syntax are not rewritten by accident.
Images and reference definition lines have their own controls because teams disagree on whether
 should become alt text, a citation, nothing at all, or stay as image syntax.
Worked example
Input:
Read [the docs](https://example.com/docs) and see .
[old-ref]: https://example.com/old
`[example](kept)`
With the default settings, the output is:
Read the docs and see diagram.
`[example](kept)`
Switch Inline links become to Text with URL and the first sentence becomes
Read the docs (https://example.com/docs) and see diagram.. Switch Images become to
Keep image Markdown and the image syntax stays untouched while normal links are still
flattened.
What it handles
- Inline links:
[label](https://example.com), including destinations wrapped in<...>and an optional quoted title after the destination. - Image syntax:
, with controls for alt text, alt text plus URL, drop, or keep Markdown. - Reference definition lines:
[id]: https://example.com, which can be dropped after the inline links are flattened or kept for review. - Code spans and fenced code blocks, preserved by default.
Limits and edge cases
- Up to 1,000,000 bytes of Markdown per run. Split larger documents and run the parts separately.
- This is a syntax flattener, not a full CommonMark renderer. It does not resolve reference-style
link uses such as
[label][id]to their URLs; it can only drop or keep the definition lines. - Nested brackets in link labels are handled, but malformed links are left exactly as written rather than guessed at.
- Relative URLs, fragments, mailto links and paths are treated as plain destinations. The tool does not fetch or validate them.
- Everything runs in your browser. No upload, no account, no network call, and no host clock.
FAQ
Will it remove the visible link text?
Only if you choose URL only. The default keeps the visible label, so [docs](https://example.com)
becomes docs. The Text with URL mode keeps both pieces as docs (https://example.com), which
is often the safest format for plain email or review notes.
What happens to images?
Images are controlled separately from normal links. By default  becomes
diagram, because alt text is the closest plain-text replacement. You can instead write
diagram (diagram.png), remove images entirely, or keep the original image Markdown unchanged.
Does it change links inside code examples?
Not by default. Backtick code spans and fenced code blocks are copied through unchanged, so a README
example like `[x](y)` remains an example. Turn Preserve code spans and fenced code blocks
off only when you intentionally want to flatten every link-looking string in the document.
Can it resolve reference-style links?
It can drop or keep reference definition lines like [docs]: https://example.com, but it does not
rewrite [the docs][docs] to the docs (https://example.com). Resolving references correctly needs
a fuller Markdown document model. This tool keeps that use visible rather than silently guessing.
Are URLs checked or fetched?
No. Destinations are treated as text. The tool does not follow redirects, validate domains, download images, or contact the network, which keeps it deterministic and safe for drafts that contain private URLs.
Developer & Automation Access
Run it from the terminal
Same engine as this page, headless — via the gizza CLI:
gizza tool markdown-link-flattener 'Read [the docs](https://example.com/docs) and see .
[old-ref]: https://example.com/old
`[example](kept)`'New to the CLI? Get gizza →
Open it by URL
Pre-fill and auto-run this tool with query parameters — the names match the API/CLI:
https://gizza.ai/tools/markdown-link-flattener/?markdown=Read%20%5Bthe%20docs%5D%28https%3A%2F%2Fexample.com%2Fdocs%29%20and%20see%20%21%5Bdiagram%5D%28diagram.png%29.%0A%0A%5Bold-ref%5D%3A%20https%3A%2F%2Fexample.com%2Fold%0A%0A%60%5Bexample%5D%28kept%29%60&link_mode=text&image_mode=alt_text&reference_definitions=drop&preserve_code=trueMachine-readable descriptor: tool.json — title + parameters JSON Schema for agents.