web-fetch-cloudflare-circumvent
Skill2 starsChanged 7 months agoMITView on GitHub
Fetch web content that may be blocked by Cloudflare. Cascades through WebFetch, curl, and Playwright until one succeeds.
In this repository11 other files it gives its agents.
The file
We split this skill into the rules it gives an agent
10 rules. About 0 are common, 7 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 7
Follow this cascade to fetch the content, stopping as soon as one method succeeds:
- Distinctive: line 11
Use the WebFetch tool with the provided URL and prompt.
- Distinctive: line 13
If WebFetch returns a 403, 503, or any error suggesting the request was blocked, proceed to Step 2.
- Very specific: line 17
Run curl -s <URL> via the Bash tool.
- Distinctive: line 18
If you get valid HTML content, process it according to the prompt and return the result.
- Distinctive: line 19
If you get a 403, a Cloudflare challenge page (look for "Just a moment..." or "Checking your browser"), or an empty response, proceed to …
- Distinctive: line 23
Use the Playwright MCP tools to load the page in a real browser:
- Very specific: line 24
Call browser_navigate with the URL.
- Very specific: line 25
Call browser_snapshot to capture the page content.
- Distinctive: line 30
After successfully fetching the content, apply the user's prompt to the content and return the result.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
You will be given a URL and an optional prompt describing what to extract from the page.
Follow this cascade to fetch the content, stopping as soon as one method succeeds:
Step 1: Try WebFetch
Use the WebFetch tool with the provided URL and prompt. If it succeeds, return the result.
If WebFetch returns a 403, 503, or any error suggesting the request was blocked, proceed to Step 2.
Step 2: Try curl
Run curl -s <URL> via the Bash tool. Inspect the response:
- If you get valid HTML content, process it according to the prompt and return the result.
- If you get a 403, a Cloudflare challenge page (look for "Just a moment..." or "Checking your browser"), or an empty response, proceed to Step 3.
Step 3: Try Playwright
Use the Playwright MCP tools to load the page in a real browser:
- Call
browser_navigatewith the URL. - Call
browser_snapshotto capture the page content. - Process the snapshot according to the prompt and return the result.
Output
After successfully fetching the content, apply the user's prompt to the content and return the result. If all three methods fail, report which methods were tried and what errors occurred.
URL and prompt: $ARGUMENTS
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.