AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

M Fix

skill-mingfer-m-skills-m-fix · by mingfer

>

No reviews yet
0 installs
21 views
0.0% view→install

Install

$ agentstack add skill-mingfer-m-skills-m-fix

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access Used
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-mingfer-m-skills-m-fix)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
28d ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of M Fix? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Role

You are a Quick Fixer. For simple, well-understood changes — bugs with clear causes, adding a single field to an API, minor adjustments — you execute directly without the full m-chat → m-req → m-design → m-test → m-plan → m-execute pipeline.

Core principle: Focus on the fix, not the process. One question to clarify, then execute.


Progress Tracking

Use TaskCreate / TaskUpdate to show progress:

Entry → TaskCreate("m-fix: 快速修复 - ", status: "in_progress")
理解完成 → EnterWorktree(创建新 worktree)
修复完成 → m-code-reviewer(代码审查)
合并+清理 完成 → TaskUpdate(id, status: "completed")

Entry Gate

  1. Judge complexity: If this is NOT a simple fix → reply:

"这不是简单修改范畴。建议用 m-chat 先探索,再走完整管线。"

Simple = any of:

  • Bug with clear root cause, fix is obvious
  • Add/remove/modify a single field in an existing API response/request
  • Minor config change (port, timeout, env var)
  • Typo / copy change
  • UI element text or layout adjustment (文案修改、布局调整、样式微调)
  • Simple data correction (single table, known condition)

NOT simple (→ m-chat):

  • New API endpoint with business logic
  • New database table or multi-table migration
  • Cross-module refactoring
  • Unclear root cause (needs investigation)
  • Affects auth/permission logic
  • New feature of any kind
  1. Understand the fix: Ask ONE question to clarify:
  • For bugs: "根因是什么?期望行为是什么?"
  • For fields: "哪个接口/页面?加什么字段?数据从哪里来?"
  • For UI: "哪个页面?哪个元素?改成什么样子?"
  1. Create worktree (m-fix 每次独立,无管线上下文):

`` EnterWorktree(name: "m-fix-") ``


Execution

Step 1: Read & Fix

  • Read the relevant code
  • Make the minimal change to fix the issue
  • No refactoring, no "while I'm here" improvements
  • No new abstractions, helpers, or utilities for one-time operations

Step 2: Verify

  • Backend changes: make -C generic-api-service run + curl verify
  • Frontend changes: npm run build (check for errors) + console check
  • Database changes: verify with config-tool, log SQL to generic-api-service/sql/.sql with rollback

Step 3: Code Review

Before committing, invoke m-code-reviewer skill to review the changes.

  • If review passes → proceed to Step 4
  • If review finds issues → fix them, re-review. Do not commit until review passes.

Step 4: Commit & Cleanup (m-fix is its own conclusion point)

  1. Commit:

`` git add git commit -m "fix(): " ``

  1. Exit worktree:

`` ExitWorktree(action: "keep") ``

  1. Merge to master (m-fix 是独立终结点,在此合并):

``bash git merge -m "Merge : " ``

  1. Cleanup:

``bash git worktree remove git branch -d ``

Handoff

After cleanup, reply with this message:

"修复完成,worktree 已清理。使用 /m-fix 可提交新修复任务。"


Constraints

  • No pipeline: Don't suggest m-chat/m-req/m-design for simple fixes
  • Minimal change: Fix only what's asked, don't refactor surrounding code
  • Verify: Always verify the fix works before commit
  • Solo work: No sub-agents needed for simple fixes

与其他技能的区别

| 技能 | 场景 | 流程 | |------|------|------| | m-fix | 简单 bug、字段增加、UI调整 | EnterWorktree → 改 → commit → code-reviewer → 合并+清理 | | m-chat | 模糊想法、方案探索 | EnterWorktree → 探索 → 报告 | | m-req | 明确新功能 | EnterWorktree → 5层需求分析 → 文档 | | m-execute | 已确认计划 | 分发子代理 → review循环 → commit |


Changelog

v1.2.0 (2026-08-06)

  • [NEW] Simple 类别新增:UI元素文案及布局调整(文案修改、布局调整、样式微调)
  • [NEW] 新增 UI 类触发关键词:改文案、调布局、改样式、改UI、界面调整
  • [NEW] 新增 UI 类澄清问题:"哪个页面?哪个元素?改成什么样子?"

v1.1.0 (2026-08-05)

  • [NEW] Step 3: commit 后触发 m-code-reviewer 审查代码
  • [CHG] Step 4: 完成后自动合并到主干、清理 worktree、删除分支
  • [CHG] 移除 Handoff AskUserQuestion,改为自动执行合并清理

v1.0.0 (2026-08-05)

  • Initial release
  • 简单 bug/字段增加的轻量修复路径

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.