Skip to content
Agent Workspaces
Skill

/md-audit

Read-only code quality audit — scan the current working directory for common issues (bugs, dead code, security hotspots, missing error handling) and return a prioritised findings report. No files are edited. Use when asked to "audit the code", "quick audit", "find issues", "code

From plugin
munder-difflin
9.5k18 skills

Claude Code

$ npx -y skills add harnessmd/munder-difflin --skill md-audit --agent claude-code

SKILL.md

md-audit.SKILL.md
name: md-audit
version: 1.0.0
description: |
  Read-only code quality audit — scan the current working directory for common
  issues (bugs, dead code, security hotspots, missing error handling) and return
  a prioritised findings report. No files are edited.
  Use when asked to "audit the code", "quick audit", "find issues", "code scan",
  or "what's wrong with this codebase". (munder-difflin)
allowed-tools:
  - Read
  - Bash
  - Grep

Code Audit (read-only)

Perform a read-only code quality scan and return a findings report. Do NOT edit any files.

Steps: 1. **Scope** — identify the primary language and entry points (`package.json`, `Cargo.toml`, `go.mod`, `pyproject.toml`, or similar). 2. **Scan** — use `Grep` and `Read` to look for:

  • Unhandled promise rejections / ignored errors
  • Hard-coded secrets or credentials (keys, tokens, passwords)
  • `TODO`/`FIXME`/`HACK` comments that signal known debt
  • Dead exports (exported symbols with no in-repo import)
  • Obvious type-safety gaps (unchecked `any`, missing null guards)

3. **Report** — output a findings list sorted by severity (Critical → High → Medium → Low):

   ## Audit Report — <project name>

   ### Critical
   - [file:line] <description>

   ### High
   - ...

   ### Summary
   <total count> findings across <file count> files scanned.

4. Stop after reporting. Do not apply any fixes.

@chaitanyagiri
Ships withmunder-difflin

an open-source alternative to the dots, bots and muses of the world, run an office of claude code/codex like agents on your laptop, sandboxes or anywhere, uses your existing subscriptions

Stars on GitHub

Momentum

stars in total
9.5k
Star history of chaitanyagiri/munder-difflin on GitHub, a chart by star-history.com

Other skills on munder-difflin.

ian-xiaohei-illustrations

ian-xiaohei-illustrati…

生成 Ian 风格的中文正文配图。用于用户要求为中文文章、帖子、博客、Notion 文档、工作流文档、方法论、流程、结构、状态、隐喻或观点生成“怪诞”“小黑”“手绘”“正文配图”“文章插图”“配图建议”“shot…

last30Days

last30Days

Resolve "last30Days" to a concrete ISO date range relative to your run time — a rolling 30-day window ending today. Returns…

last7Days

last7Days

Resolve "last7Days" to a concrete ISO date range relative to your run time — a rolling 7-day window ending today. Returns…

lastMonth

lastMonth

Resolve "lastMonth" to a concrete ISO date range relative to your run time — the previous full calendar month. Returns inclusive…

lastQuarter

lastQuarter

Resolve "lastQuarter" to a concrete ISO date range relative to your run time — the previous full quarter. Returns inclusive civil…

More skills from other plugins.

paseo

paseo

Paseo reference for managing hosts, projects, workspaces, workspace scripts, agents, schedules, and heartbeats.

paseo10d23k
@getpaseo@getpaseoInstall Skill
10x
Auto-invoked

10x

Personalized audit that teaches the advanced Paradev (formerly Superset) features the user isn't using yet (automations, parallel…

paradev1d16k
@superset-sh@superset-shInstall Skill
brainstorming

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or…

codeg1d4.4k
@xintaofei@xintaofeiInstall Skill

Questions about md-audit.

Which coding agents can use md-audit?
md-audit installs in Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode and Windsurf with the skills CLI, using each agent's flag shown on this page. Flowy's auto-invocation, which fires a skill when your prompt matches, runs in Claude Code only.