A pattern-based community killfile for outgoing Hubot messages. Implements a middleware to intercept messages Hubot is about to post, refrains from outputting matched messages.
Perhaps you'd like to prevent Hubot being trained to swear, or you use hubot-twitter-mention but you want to filter out some returned Tweets.
Uses robot.brain to persist killfile entries.
All configuration is done using commands as below.
hubot refrain (on|off) - Activate/deactivate refrain filter.
hubot refrain show - Print the list of patterns to refrain from saying.
hubot refrain add <pattern> - Add refrain filter entry
hubot refrain remove <N> - Delete refrain filter entry (by index)