[BUG] Webfetch not working in cloud version
Status Fixed / completed
Maintainer reply None cached
Workaround ✓ Mentioned in thread ↓
Activity 4 comments · opened Dec 11, 2025 · closed Jan 22, 2026
Preflight Checklist
- [x] I have searched existing issues and this hasn't been reported yet
- [x] This is a single bug report (please file separate reports for different bugs)
- [x] I am using the latest version of Claude Code
What's Wrong?
WebFetch not working in cloud version - All WebFetch requests return 403 Forbidden errors
What Should Happen?
WebFetch should successfully retrieve content from publicly accessible URLs like example.com
Error Messages/Logs
403Steps to Reproduce
Use WebFetch tool with any URL (including https://example.com)
- Receive "Request failed with status code 403"
- WebSearch tool works fine in same environment
Claude Model
Not sure / Multiple models
Is this a regression?
I don't know
Last Working Version
_No response_
Claude Code Version
cloud
Platform
Anthropic API
Operating System
Other
Terminal/Shell
Non-interactive/CI environment
Additional Information
Environment details:
Using Claude Code in cloud/web version
Session ID: 01Jd7zfmR3Jg8of9KWWV9M7i (from this session)
Date: 2025-12-11
4 Comments
This issue has been inactive for 30 days. If the issue is still occurring, please comment to let us know. Otherwise, this issue will be automatically closed in 30 days for housekeeping purposes.
@briangflett Are you still seeing this issue? I'm encountering this issue frequently during research and discovery tasks.
Problem (According to Google):
Claude Code's web fetching features frequently encounter 403 Forbidden errors because many websites actively block automated access from AI agents or bots. The 403 error means the server understood the request but refuses to fulfill it, often due to insufficient permissions or server-level security settings that detect and block non-human traffic.
This is a delicate problem
If we don't handle this proactively this going to create a TON of unnecessary repercussions for the open web. I unterstand the implications of too much autonomous demand costing hosts. However, I'm confident many content providers would appreciate an alternative to a hard 403 error. Furthermore, all of the solutions below slow down e/acc. We need a better way...
I'm curious about configuring a universal cache, similar to what Google used to disclose under each result.
---
Solutions and Workarounds (From Google)
Manual Compilation/Local Fetching: For coding projects, users may need to manually fetch or compile resources on their local machine and provide the content to Claude, bypassing the environment's web fetch limitations.
Proxy Configuration: Ensuring the correct HTTP proxy settings are used and that the ALL_PROXY environment variable is unset can resolve certain internal authentication-related 403 errors with the CLI tool.
Provide Content Directly: If a URL consistently returns a 403, users can copy the content and paste it into the chat interface for the AI to analyze, rather than relying on the automated web fetch tool.
@m31uk3 the problem is not happening as often. I am now able to use the webfetch tool to get example.com and specific files in my github repository. I'll close this for now, and open another later when I come upon another specific example.
This issue has been automatically locked since it was closed and has not had any activity for 7 days. If you're experiencing a similar issue, please file a new issue and reference this one if it's relevant.