HTTP Error 403: Forbidden

[codemurt]

I'm trying to run this code:

from claude import claude_client
from claude import claude_wrapper
SESSION_KEY = <MY_SESSION_KEY>
client = claude_client.ClaudeClient(SESSION_KEY)
organizations = client.get_organizations()
organizations

but I get this:

WARNING:root:Failed response object: Response(ok=False, data=b'', status_code=None, error='HTTP Error 403: Forbidden')
None

I don't really understand what the problem is.

[AshwinPathi]

Could you dump the logs, try a different session key, or try a different account?

[sakurakoujirabbit]

I have the same problem. [17-Mar-24 18:10:48:root:WARNING][claude_client.py:304 - get_organizations() ] Failed response object: Response(ok=False, data=b'', status_code=None, error='HTTP Error 403: Forbidden') None

[gerkim62]

same case here

[jensen1207]

same error that happened since last Friday

[AshwinPathi]

same, probably some additional headers or security was added recently.

[AshwinPathi]

@jensen1207 @gerkim62 @sakurakoujirabbit @codemurt

The problem is anthropic uses cloudflare cookies to prevent bots. These cookies refresh periodically, so this probably isn't a permanent solution. The "real" best way to solve this would be to completely mock the browser or find some way to bypass cloudflare cookies. The former is too heavyweight and the latter might be difficult.

Since this is also kind of against their TOS, I don't think I can comment more.

[Kneegrowtreegrow]

User Cookie_browser3

import browser_cookie3

cj = browser_cookie3.chrome(domain_name='claude.ai') cookies = {cookie.name: cookie.value for cookie in cj}

print(cj)

print(cookies)

You can get all the present cookies from the chrome browser even the cloudflare cookies.