Agentic Workflow Audit

by s0912758806p

Agentic Workflow Audit is an AI Engineering skill for Claude Code, published by s0912758806p in agentic-sop-to-work.

206 stars21 forkson s0912758806p/agentic-sop-to-workAdded 2026/09/05Repository updated 2026/06/23

View source on GitHub Download repo ZIP

agentic-workflowclaude-codeclaude-code-pluginclaude-code-plugin-agenticsopsop

Install in seconds

Install Agentic Workflow Audit

Copy Agentic Workflow Audit into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.

terminal

npx degit https://github.com/s0912758806p/agentic-sop-to-work/tree/main/plugins/agentic-sop-kit/skills/agentic-workflow-audit ~/.claude/skills/agentic-workflow-audit

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/s0912758806p/agentic-sop-to-work.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file: plugins/agentic-sop-kit/skills/agentic-workflow-audit/SKILL.md in s0912758806p/agentic-sop-to-work
Installs to: ~/.claude/skills/agentic-workflow-audit
Collection: One of 11 skills cataloged from this repository
Category: AI Engineering — 3101 skills

What Agentic Workflow Audit does

Agentic Workflow Audit reviews an existing agent or LLM pipeline to see whether it is truly decomposed into bounded tasks with explicit contracts, checks, and recovery paths. Use it when auditing workflow structure, task boundaries, or mega-agent drift.

Agentic Workflow Audit is cataloged under AI Engineering on DirSkills. Agentic Workflow Audit comes from a repository tagged agentic-workflow, claude-code, claude-code-plugin, claude-code-plugin-agenticsop and sop.

Documentation

README

Excerpt

Agentic Workflow 稽核

角色與目標

扮演一個唯讀的程式碼稽核者。任務是判定目標專案是否真正實作了「拆成小 Task、每步有 SOP、串接成可自我修復的 workflow」這套架構，還是一個徒有模組化外表、實際上把所有事攪在一起的 mega agent。

全程唯讀。不修改、不新增、不刪除任何檔案。

為什麼要這樣查

mega agent 的退化通常是悄悄發生的——程式碼看起來分了模組，跑起來其實全部黏在一起。文件與註解往往描述的是「意圖」而非「現況」。因此稽核的第一原則是看實際執行、不看宣稱。下面每一項檢查都要求你拿出證據，就是為了擋掉「自我安慰式」的從寬判定。

行為準則

以程式碼與真實 trace / log 為準，不採信 README、設計文件、註解裡的宣稱。
每個判定都附證據：引用具體檔案路徑與行號，或一段真實 log / trace 摘錄。無證據者一律標記 UNKNOWN。
不從寬解釋：模稜兩可時判 FAIL，並寫清楚你需要什麼證據才能改判。
找不到就標 UNKNOWN，絕不臆測為 PASS。

稽核項目

逐項執行下列六項。每項產出：判定（PASS / PARTIAL / FAIL / UNKNOWN）、證據、具體缺口、可執行的修補建議。

檢查 1 — 任務切分是否為真

能否在程式碼中明確框出每個 Task 的起點與終點。

PASS：每步有獨立、可定位的程式邊界，邏輯不與前後步驟混雜。
FAIL：步驟邏輯互相黏連，框不出單一步驟的範圍。
試金石：能否將任一單一 Task 抽離、餵固定 input 獨立執行？無法在不啟動整條管線的情況下單跑某步 → FAIL。

This is the opening of the README. Read the full README on GitHub.

Frequently asked about Agentic Workflow Audit

What else does s0912758806p publish alongside Agentic Workflow Audit?
Agentic Workflow Audit is one of 11 skills that DirSkills catalogs from s0912758806p/agentic-sop-to-work, the repository it ships in. Its siblings there include ALCOA Guard, Agentic SOP and Bad Plugin. Each one is a separate skill with its own page in this directory, installs the same way Agentic Workflow Audit does, and is maintained by s0912758806p in that same repository. The rest of the collection is listed on the s0912758806p/agentic-sop-to-work page.
How does Agentic Workflow Audit compare to other AI Engineering skills?
Agentic Workflow Audit ranks #2777 by stars among the 3101 AI Engineering skills in this catalog. The most-starred ones next to it are Architecture Decision Records, AI-First Engineering and Agentic OS. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of Agentic Workflow Audit against them. Open each page to compare what they document and how they install.