Context
AbuseFilter provides the ability to define consequences in response to triggers. These are currently things like block, deny action, etc. It would be nice to define a consequence to show a CAPTCHA via the ConfirmEdit extension, for cases where the AbuseFilter exists to deny bot editing.
Proposal
- Introduce a CaptchaConsequence that can be used for actions in AbuseFilter
- Users with skipcaptcha right can continue to bypass CAPTCHA even if they trigger the AbuseFilter
Consequences
- It is possible to configure AbuseFilters that show a CAPTCHA when conditions are matched
Next steps
- Complete QA
- Enable on all wikis
Original report
Author: mike.lifeguard+bugs
Description:
To either slow down human abusers or weed out bots, please allow AbuseFilter to present a CAPTCHA & allow the change only if solved. Yes, I'm aware that captchas are mean to humans both blind and sighted (though of course a particular horror from an accessibility standpoint) - they'd have to be used sparingly etc, but that's a useful option IMO.
Version: unspecified
Severity: enhancement