Event ·Multi-source

工程师分享：Tag 每天完成超 50% 的 PR 和全部数据分析

一位工程师在 X 上分享日常用法：Claude 驱动的 Tag 每天替他写超过 50% 的 PR，数据分析约占 100%，并自动修复多数产品反馈和 bug。搭配 Opus 5.5 与 Fable 5.1 模型，Tag 具备主动性、记忆和连接器访问能力。他给出 4 条实际提示词，包括让 Claude 在频道里复现 bug 并提 PR、用约 100 个假设和约 1000 万 token 深挖异常数据、生成代码讲解游戏和幻灯片。

Current read

有人把 Claude 接进 Slack 当团队自动化员工用，复现 bug 自动提 PR、花千万 token 挖数据，4 条提示词直接抄走。

11 sources83° AI heatLast updated 2026/9/25 17:34:34

Evidence

11 sources

Primary sourceBoris Cherny

2026/9/25 17:34:34View source

Related source 1Claude

2026/9/23 19:21:42View source

Related source 2TestingCatalog

2026/9/24 06:55:33View source

Related source 3IT之家

2026/9/24 23:51:55View source

Related source 4AIGCLINK

2026/9/25 01:55:43View source

Related source 5Shubham Saboo

2026/9/25 02:28:54View source

Related source 6HeyGen

2026/9/25 18:03:28View source

Related source 7kimmonismus

2026/9/26 10:22:51View source

Related source 8Vaibhav Srivastav

2026/9/23 18:08:08View source

Related source 9rohanpaul_ai

2026/9/23 18:44:00View source

Related source 10Notion

2026/9/23 20:33:25View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story