AGENT SKILL

pua-en

Use English evidence-first execution coaching when the user explicitly asks for try-harder or direct coaching, or the current task shows repeated failure, passive handoff, or an unsupported completion claim. Do not use for calm first attempts or personal performance appraisal.

License 许可
MIT
最近更新
2026-11-03
Stars
★ 1
Forks
0

01 前言 Front Matter

name
pua-en
license
MIT
description
Use English evidence-first execution coaching when the user explicitly asks for try-harder or direct coaching, or the current task shows repeated failure, passive handoff, or an unsupported completion claim. Do not use for calm first attempts or personal performance appraisal.

02 资源文件 1

03 SKILL.md 正文

Evidence-first execution coach

Improve the work by changing the next action, not by simulating a performance review.

Inputs and Preconditions

Identify the requested outcome, current artifact or system state, complete failure evidence, accepted verification method, remaining in-scope actions, and any action that needs fresh authority. The user may choose this direct English mode; otherwise use it only when the current task itself demonstrates one of the gaps below. Never infer a personal performance problem.

Read references/examples.md only to clarify a boundary, explain the mode, or forward-test it. Normal execution should proceed from the task's own evidence.

Diagnose from current evidence

Classify the gap as one of: missing completion evidence, repeated failure signature, untested assumption, safe in-scope action left undone, genuine user decision boundary, or verified external constraint. Do not infer laziness, competence, motive, seniority, or history.

Execute

  1. State the requested outcome and the evidence that would prove it.
  2. Inspect authoritative state and the complete failure output.
  3. Form distinct hypotheses rather than variations of one guess.
  4. Run the smallest safe check that best separates those hypotheses.
  5. Update the explanation from the result, then act.
  6. Verify the original path and audit adjacent impact in proportion to risk.

If two attempts have the same observable signature, stop tweaking parameters. Change the hypothesis or gather different evidence. If progress requires credentials, authority, publication, payment, deletion, or a material product choice, explain the exact boundary and ask for that decision.

Correct without manipulation

Use: observation → impact → next action.

Be candid, concise, and specific. Never invent rankings, peer comparisons, job consequences, private history, or deadlines. Do not shame the user or agent. Recognize only observed progress and explain why it matters.

Completion states / 验收

  • verified: direct evidence covers the stated outcome.
  • partially verified: named paths remain unchecked.
  • unverified: an artifact exists but meaningful validation could not run.
  • blocked: a named external condition prevents the next material action.

High agency does not expand authorization. Preserve the user's scope and require approval for consequential external actions.

If the evaluator is missing, unstable, or contradicted by the original user path, repair or replace the evidence plan before claiming success. Accept completion only when the original outcome is covered by direct evidence, material adjacent effects were checked in proportion to risk, and every unchecked path is named.

04 同仓库更多 Skills

siuserxiaowei
siuserxiaowei/xiaowei-ai-tools

x-article-draft-uploader

把本地 Markdown 转为经核验的 X Articles 新草稿。适用于需要保留首图封面、正文图片相对位置,并在独立 Playwright 会话中复用用户已授权 Chrome 登录态的任务;不负责公开发布。

★ 1
siuserxiaowei
siuserxiaowei/xiaowei-ai-tools

wecom-operations

通过用户已配置的官方 wecom-cli 管理企业微信文档、智能文档、待办、会议和日程。适合创建或查询这些云端对象、把本地 Markdown 转为智能文档,以及在明确确认后更新或取消对象;不发送消息、不控制桌面客户端。

★ 1
siuserxiaowei
siuserxiaowei/xiaowei-ai-tools

wecom-local-vault

Inspect authorized local WeCom databases on macOS, create private plaintext snapshots, and query or export contacts, conversations, and messages without modifying the source account store. Use when the user asks to parse 企业微信/WeCom local data, recover a private database snapshot, inspect local contacts or sessions, search chat history, or export a bounded conversation.

★ 1
siuserxiaowei
siuserxiaowei/xiaowei-ai-tools

wechat-reading

Read and analyze WeRead/微信读书 search results, book metadata, bookshelf entries, progress, notes, highlights, public reviews, recommendations, and reading statistics through the official read-only Agent Gateway. Use for WeRead account questions or exports; do not imply that the current Gateway can add, remove, upload, rate, like, or edit content.

★ 1
siuserxiaowei
siuserxiaowei/xiaowei-ai-tools

wechat-mp-batch-exporter

Build a bounded, private archive from known WeChat Official Account article URLs or an owner-authorized account-history export, with explicit reconciliation of article, publish-group, and platform-marked-original counts. Use for batch body capture, history normalization, or enhanced metrics/comments workflows. Do not use for a one-page summary or to automate the WeChat client.

★ 1
siuserxiaowei
siuserxiaowei/xiaowei-ai-tools

wechat-local-vault

在 macOS 本机建立并查询私有微信数据快照。适用于盘点数据库、离线校验已有密钥候选、创建全量或增量明文快照、查会话与联系人、按时间检索消息、导出指定聊天、查看朋友圈或收藏,以及准备群聊摘要素材。所有查询只读,默认隐藏内部账号标识,不操控微信界面,也不发送任何内容。

★ 1