Repository navigation
feat: 个性化动态称谓、持续信任积累与控制台优化(0.7.0) - #59
Conversation
Codex Review Summary: ✅ CompletedThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: d112e9aad9
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| for group, user, reward in { | ||
| (group, user, reward) for group, user, _, reward in affected if reward | ||
| }: |
There was a problem hiding this comment.
Reverse non-boundary trust rewards when their sources are revoked
When a promise_kept, verified_correction, or repair_action event references a persisted boundary anchor, revoking that anchor invalidates the dependent evidence but leaves its trust reward intact. These events never populate consumed_by, so filtering the reversal set to nonempty reward values excludes their own scored logs. I reproduced this through _snapshot(), verify_trust_evidence(), and FavorService.undo_log(): a dependent promise_kept awards +0.5 T, and undoing its source leaves that +0.5 T despite both evidence rows becoming invalidated. Include the scored dependent events themselves in the reversal set, while deduplicating boundary batch rewards, so revoked evidence cannot continue contributing trust.
AGENTS.md reference: AGENTS.md:L13-L13
Useful? React with 👍 / 👎.
本次更新让称谓能体现 Bot 的人格,为持续可靠的行为补充信任奖励,并改善管理页面的状态显示。版本更新为 0.7.0。
新安装默认在回复中体现关系;已有明确关闭的选择继续保留。升级保留关系分数和历史流水,自动兼容旧称谓设置。备份和回退方法见仓库中的
docs/MAINTENANCE_UPGRADE.md。功能验证已通过:Python 1784 项、Node 100 项,以及 AstrBot 4.28.0 / 4.29.4 的十组框架检查;Python 另有 1 项跳过。本次文档修订另行检查版本、链接、公开内容和格式。真实群聊、线上升级与联网模型的表达效果仍需实际使用验证,完整记录见
docs/ACCEPTANCE.md。