Turn rough software ideas into bounded MVP scopes, practical architecture decisions, and coding-agent-ready implementation tasks without unnecessary complexity.
DriftGuard analyzes a repository for dependency and API upgrade impact, identifies affected code, applies targeted migration changes, and validates the result with tests. It returns a structured migration result with changed files, diff, migration report, test results, and PR-ready summary. Runs with Claude Agent in Capafy and supports terminal/CI workflows.
Forces AI coding agents to read exports, callers, and shared utilities before touching any file — then restricts every edit to the declared scope, flags conflicting patterns for human resolution, and reports any file touched outside the boundary.
Compare your SOW with client requests, expose margin-eroding scope creep, and get a ready-to-send Change Order pack.
你跟 AI 說「幫我規劃一個 X」，它兩秒就吐出一份看起來無懈可擊的計畫——問題是，裡面一半假設是它自己掰的，你分不出是哪一半。照著做，兩天後才發現方向從第一步就錯了，全部重來。這支反過來做：先把你腦中真正在想的東西窮追猛打地挖出來，逼你自己想清楚，在你確認達成共識之前，它一個字都不動。四條硬規則讓它不惹人厭：每題都附推薦答案不丟空白問卷、事實自己查只問真正該你決定的、一次只問一題不讓你腦當機、沒共識就不開工。附 18 個追問鏡頭——被否決的替代方案、機會成本、這是誰的渴望、Pre-mortem。
编程专家.Skill P8级编程助手,覆盖：软件/网站项目总控、API设计、Bug诊断、代码生成、代码审查、重构、测试用例、性能基准、技术选型、文档生成、任务拆解、Spec驱动、Karpathy规范、CMS二次开发、前端设计、MySQL、项目知识图谱。支持 @标识 显式调用跳过路由。
Audit RFP and tender packages, trace every requirement to its source, flag proposal gaps, and generate a compliance matrix, prioritized gap report, and submission checklist.
在各分項工程設檢驗停留點（限止點），引導廠商自主檢查後提出查驗申請、監造會同查驗簽認，產出各工項施工抽查標準表與紀錄表（測量放樣、鋼筋、模板、混凝土、結構鋼材、機電），並辦理不符合事項的缺失改善追蹤與統計。依《公共工程施工品質管理作業要點》第二級品管辦理。