Defuddle
Pull clean markdown from a normal web page URL so agents read articles and docs with fewer navigation tokens than raw fetch.
Overview
Defuddle is an agent skill most often used in Build→docs (also Idea→discover and Grow→content) that extracts clean markdown from web pages via the Defuddle CLI to reduce token clutter versus raw page fetch.
Install
npx skills add https://github.com/kepano/obsidian-skills --skill defuddleWhat is this skill?
- CLI: defuddle parse <url> --md for markdown output
- Optional -o content.md save and -p title|description|domain metadata extraction
- Use instead of WebFetch for standard web pages; skip URLs already ending in .md
- Global install via npm install -g defuddle when missing
- Supports --json for combined HTML and markdown when needed
Adoption & trust: 29k installs on skills.sh; 34.9k GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
You gave your agent a documentation or article URL and WebFetch floods context with nav, ads, and chrome.
Who is it for?
Solo builders and note-takers who routinely ingest public web documentation, blog posts, or help articles into agent or Obsidian workflows.
Skip if: Raw .md file URLs (use WebFetch), authenticated paywalls without another auth path, or tasks needing full rendered DOM interaction.
When should I use this skill?
User provides a URL to read or analyze for online documentation, articles, or standard web pages—not URLs ending in .md.
What do I get? / Deliverables
You get focused markdown (or saved content.md) suitable for analysis, note-taking, or spec drafting with lower token overhead.
- Markdown page content in agent context
- Optional content.md file on disk
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Canonical shelf is Build→docs because the primary trigger is reading online documentation and articles while implementing or documenting. Defuddle strips clutter from HTML pages into markdown—the same moment you ingest external docs into notes, specs, or agent context.
Where it fits
Pull vendor API guide markdown into your repo README without sidebar HTML noise.
Read a competitor launch post as clean markdown for a positioning memo.
Harvest article body text before drafting a newsletter or changelog summary.
How it compares
CLI content extractor for standard pages—not a headless browser skill and not WebFetch for already-markdown URLs.
Common Questions / FAQ
Who is defuddle for?
Claude Code, Cursor, and Codex users who want cleaner page text when researching or documenting from public HTTP URLs.
When should I use defuddle?
During Build docs work when citing vendor guides, during Idea discover when reading competitor articles, and during Grow content research when harvesting readable source material.
Is defuddle safe to install?
Check Prism Security Audits for the Obsidian skills package; the skill runs npm global install and fetches arbitrary user-supplied URLs over the network.
SKILL.md
READMESKILL.md - Defuddle
# Defuddle Use Defuddle CLI to extract clean readable content from web pages. Prefer over WebFetch for standard web pages — it removes navigation, ads, and clutter, reducing token usage. If not installed: `npm install -g defuddle` ## Usage Always use `--md` for markdown output: ```bash defuddle parse <url> --md ``` Save to file: ```bash defuddle parse <url> --md -o content.md ``` Extract specific metadata: ```bash defuddle parse <url> -p title defuddle parse <url> -p description defuddle parse <url> -p domain ``` ## Output formats | Flag | Format | |------|--------| | `--md` | Markdown (default choice) | | `--json` | JSON with both HTML and markdown | | (none) | HTML | | `-p <name>` | Specific metadata property |