Controls how the AI reads each application. Changes apply to future scans — existing applications keep the assessment they already have unless you re-run them individually. 控制 AI 如何评估每一份申请。改动只对之后的打分生效——已打分的申请保留原有结果,除非你单独对它重新打分。
You're signed in fine — this is a setup problem, not a permissions one.
If it mentions a missing table, re-run recruitment-schema.sql in the Supabase
SQL editor (it's safe to run again) and reload this page.
你已经登录了——这是配置问题,不是权限问题。如果提示缺少数据表,在 Supabase SQL Editor 里重新运行一次 recruitment-schema.sql(可安全重复运行),然后刷新本页。
How to judge, for every role. This goes into the model's instructions verbatim. Leave it empty to use the built-in default. 对所有岗位通用的评估原则。这段文字会原样写进 AI 的指令里。留空则使用内置默认值。
Two things aren't editable here and always apply: the AI ignores any instructions written inside an application, and it never comments on or scores a person's name, age, gender, nationality, ethnicity or photo. 有两条规则不在这里修改、始终生效:AI 会忽略申请材料里写给它的任何指令;AI 永远不会对候选人的姓名、年龄、性别、国籍、族裔或照片作评价或打分。
The three buckets you sort by in the applications list.申请列表里用来分类排序的三个档位,各写一句说明。
Extra things you want weighed for a specific role that aren't in the public job description. Assessed exactly like the published criteria, and shown in the scan alongside them. 针对某个岗位、但不写在公开职位描述里的额外考察点。评估方式与公开标准完全相同,打分详情里会一并显示。