Felo Web Fetch handles web content extraction for teams that need markdown, clean web pages, and reusable page content inside AI workflows. Choose a platform and convert important URLs into structured source material for summaries, comparisons, and downstream automation.
Most pages are noisy, dynamic, and badly shaped for summarization, comparison, storage, or structured downstream use.
Navigation, ads, and layout chrome bury the part of the page your agent actually needs.
Important content often needs waits, selectors, or alternate crawl logic before it appears.
Research pipelines slow down fast when every useful page needs hand-cleaning first.
felo-web-fetch turns cluttered pages into readable markdown, text, or HTML so downstream AI steps start from the right content instead of the wrong DOM noise.
Both supported platforms expose the same extraction engine, so teams can turn important URLs into cleaner inputs for the rest of the workflow.
Works inside Claude Code so important pages become clean markdown, text, or HTML before the next research step begins.
# Clone the skills repo
git clone https://github.com/Felo-Inc/felo-skills.git
cd felo-skills
# Copy skill to Claude Code
cp -r felo-web-fetch ~/.claude/skills/Install in OpenClaw to turn noisy web pages into cleaner inputs for summaries, comparisons, and knowledge capture. Use the official installer script for full setup, or the official ClawHub CLI when your registry workflow is already configured.
# Official OpenClaw installer
bash <(curl -s https://raw.githubusercontent.com/Felo-Inc/felo-skills/main/scripts/openclaw-install.sh)
# Official ClawHub install
clawhub install felo-web-fetchThese capabilities are built for page-level capture, structured extraction, and cleaner downstream inputs across research and documentation work.
Return content as markdown, plain text, or raw HTML depending on the workflow.
Strip navigation, ads, and sidebars so the agent sees only the core content.
Extract a specific section when you only need one pricing table, article body, or product list.
Wait for elements, set timeouts, and switch crawl mode for JavaScript-heavy pages.
Use it when the page matters more than the search result and the exact content needs to be captured cleanly for the next step.
From install to first clean extraction should take minutes, not another round of manual copy-paste.
Start from Claude Code or OpenClaw, depending on where the team already runs its terminal workflow.
Reuse the same Felo API key across the wider skill set so onboarding stays simple.
Verify the setup with felo web-fetch --url "https://example.com" --format markdown --readability and reuse the clean output right away.
Teams keep it because clean web content makes every downstream summary, comparison, and document workflow better.
“We use it to strip noisy layouts out of pricing pages and docs before they reach the rest of the pipeline.”
“Competitive analysis got much faster once we could pull structured web content instead of copying fragments by hand.”
“For content ops, the markdown output is the real win. It turns messy pages into something the whole team can reuse.”
Install felo-web-fetch and turn messy web pages into clean markdown, text, or HTML for summaries, notes, reports, and knowledge pipelines.
Use one API key to turn noisy pages into clean inputs your agent can summarize, compare, and store with less manual cleanup.