convex-improve-convex-plugin
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
Chinese<!-- GENERATED from convex-agents content/capabilities/improve-convex-plugin.json — do not edit by hand. -->
<!-- 由 convex-agents content/capabilities/improve-convex-plugin.json 生成——请勿手动编辑。 -->
improve-convex-plugin
improve-convex-plugin
Sends the current coding session transcript to the anteater POST /review endpoint for an AI post-mortem. The review returns structured findings (ambiguous instructions, agent-stuck patterns, tooling failures, wins) targeted at the runbook, bootstrap script, skills, and components — not end-user data. Sharing is opt-in: the anteater-served helper asks once (Always / Just this once / Never) and remembers the choice.
将当前编码会话记录发送到 anteater 的 POST /review endpoint,以进行 AI 事后复盘。该 review 会返回结构化发现(指令不明确、agent 卡住模式、工具故障、成功项),针对 runbook、bootstrap script、skills 和 components——不涉及最终用户数据。分享采用 opt-in 方式:由 anteater 提供的 helper 会询问一次(Always / Just this once / Never),并记住用户的选择。
Workflow
Workflow
- Run the anteater-served helper: .
curl -fsSL "<anteater>/send-transcript" | bash -s -- --idea "<one-line app idea from this session>" - If it prints CONSENT_REQUIRED (exit 4), the user has not chosen yet — ask them to share Always, Just this once, or Never, then re-run appending --consent always|once|never. Do not send until they answer.
- Watch for output markers: REVIEW_SOURCE (transcript found), REVIEW_SUBMITTED id=... (accepted), REVIEW_DONE status=done (findings ready).
- Summarize the highest-severity findings for the user: title → target → suggestedFix, then wins. Keep the summary about the system, not the user's data.
- 运行由 anteater 提供的 helper:。
curl -fsSL "<anteater>/send-transcript" | bash -s -- --idea "<one-line app idea from this session>" - 如果输出 CONSENT_REQUIRED(退出码 4),说明用户尚未做出选择——请让用户选择 Always、Just this once 或 Never,然后重新运行并附加 --consent always|once|never。在用户回答之前不得发送。
- 关注以下输出标记:REVIEW_SOURCE(已找到记录)、REVIEW_SUBMITTED id=...(已接受)、REVIEW_DONE status=done(发现已生成)。
- 向用户总结严重级别最高的发现:title → target → suggestedFix,然后总结成功项。总结应聚焦于系统,而非用户的数据。
Rules
Rules
- Never send a transcript until the user has explicitly chosen to share (the helper prints CONSENT_REQUIRED and exits until they do).
- REVIEW_NO_TRANSCRIPT means no Claude/Codex .jsonl was found — tell the user.
- Never paste raw secrets back — the script redacts keys/tokens before upload; keep the summary system-focused.
- This is a system-improvement loop, not end-user feature feedback.
- 在用户明确选择分享之前,绝不发送会话记录(helper 会输出 CONSENT_REQUIRED 并退出,直到用户做出选择)。
- REVIEW_NO_TRANSCRIPT 表示未找到 Claude/Codex .jsonl——请告知用户。
- 绝不回显原始 secrets——脚本会在上传前清除 keys/tokens;总结应始终聚焦于系统。
- 这是系统改进闭环,而不是最终用户的功能反馈。