Patch-Notes Composer

Paste changelog sludge or fetch a URL → get clean Markdown buckets you can ship.

Source

Markdown


          

Buckets

What this does

Turns long changelogs or blog posts into clean, skimmable buckets: Added, Changed, Fixed, Removed, Deprecated, Known issues, Open questions, and Highlights.

Quick start

  1. Fetch a URL or paste raw text.
  2. Hit Summarize. We parse headings + sentences and auto-bucket by verbs (“added”, “fixed”, “deprecated”…).
  3. Copy the result as Markdown or Export JSON for your notes/release process.

Buckets, at a glance

  • Added — new stuff, launches, newly supported flags.
  • Changed — updates, tweaks, reworks, refactors.
  • Fixed — bugs squashed, regressions addressed.
  • Removed/Deprecated — gone now or going soon.
  • Known issues — admitted limits/bugs.
  • Open questions — TBDs, “needs investigation”.
  • Highlights — imperative takeaways (“Use/Disable/Set/Stop…”) and clean headings that don’t match others.

Pro tips

  • Give it signal: paste the real changelog, not the marketing blog.
  • Version context: include a range like v1.8 → v1.10 in your paste for clarity.
  • Trim noise: if a bucket looks vague, edit in place after you paste to Markdown.
  • Short input? We’ll still draft sane bullets so you’re not left with nothing—clean them up as needed.

Export & handoff

Use Copy Markdown for PR descriptions, internal posts, or customer notes. Use Export JSON to pipe the buckets into your docs pipeline.

Nerd notes
  • Headings are treated as strong signals; near-duplicate lines are de-deduped fuzzily.
  • Fetcher has SSRF guards, a small rate-limit, and a 10-minute cache. Very large pages are truncated.