Kimi Reaches Settlement with Open Source Community: Risk Control Upgrade Accidentally Hurts Third Party Tools, 'Double Standard PR' Misunderstandings According to Beating monitoring, the Kimi Code team under Moon's Dark Side reached a settlement with the open-source community on the evening of May 26th, completely quelling the subscription risk control and public relations crisis caused by the account suspension incident. Open source developer Leechaul announced the removal of previously excessive comments and confirmed that the pi provider kimi code plugin will continue to be maintained. Paid subscription members can continue to securely use Kimi Code Plan in third-party tools such as Pi Coding Agent in the future. Kimi Code team member Young released an official Chinese statement explaining the technical details of account suspension defense. The team has recently upgraded its risk control defense strategy to combat sales and abnormal requests at transit stations, and has included the User Agent identifier as one of the criteria for judgment. However, the risk control model did not fully consider the existence of many third-party Code Agents (such as Pi) constructed as Kimi CLI User Agents in the community, resulting in the system determining normal requests as malicious forgery, which inadvertently harmed a large number of compliant subscribers. The double standard dispute is actually a misunderstanding caused by the time difference in multi-channel communication. Young explained to this channel that due to the fact that domestic big model developers mainly gather in the official Feishu group on the dark side of the moon, the team had already provided answers and appeasement within Feishu as soon as the incident occurred. However, the English response from X platform was delayed due to the time difference and cross platform operation, inadvertently creating the double label illusion of only apologizing to the outside world. At present, the dark side of the moon has comprehensively adjusted its risk control strategy, and all restricted accounts have been restored to normal. It has also started actively contacting third-party developers to optimize the User Agent judgment. After reaching a settlement, Leechaul revealed that Kimi's R&D team had had in-depth communication with Leechaul to rebuild basic confidence in the domestic large model ecosystem, and began writing exclusive prefix cache optimization guidelines for Kimi's personal intelligent agents. [Original link]