kevinshi3200/persona-auditor

kevinshi3200★ 1最后同步: 2026-08-22

在 GitHub 打开

Persona Auditor — audit AI-generated code from a real user's perspective. Digital personas exhaustively traverse every user journey, then god's-eye attribution collapses bugs into a few root causes for precise fixing. Agent skill (Claude Code / Codex / Cursor / DSH).

README 摘要

Persona Auditor Find the bugs real users hit that you'd only discover through costly trial-and-error — or never notice at all — and collapse them into a handful of root causes, so your agent fixes precisely instead of patching blindly. 从真实用户视角审计 AI 代码,找出你反复试验才能发现、甚至永远发现不了的错误,归因成少数根因, 一次修对 。 TL;DR The Problem AI writes code that runs fine and is internally consistent — but the moment a real user walks the journey, it falls apart: - Users get stuck in the first few steps , even though every function looks correct in isolation. - You patch one point after another, for many rounds, and never find the root cause — because each fix only sees a local symptom. - Worst of all: some bugs you'd never notice at all. Cross-project contamination, stale state, wrong sort order — they hide deep in the code and only surface when a real user walks the full path . The Solution Persona Auditor dispatches "digital personas" — subagents that each play a real user type — and has them exhaustively traverse every user path on paper (no execution, no clicking: cheaper and more global than real clicking). Then it steps back to a god's-eye view and attributes dozens of symptoms to a few root causes , each wit…

在 GitHub 查看完整 README →
Agent/智能体agent-skillsai-auditbug-huntingclaude-skillscode-auditcode-reviewdsh-pluginpersona

分类