How do I install the Issue skill?

Run npx skillmds@latest add mike-diff/issue-3 in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Issue safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Issue?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Issue free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

mike-diff (@mike-diff) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Issue by mike-diff — Agent Skill

Name: Issue
Author: mike-diff

File contents

/agent-team:issue - GitHub Issue Analysis & Planning

Analyze a GitHub issue and prepare an implementation plan. Research, explore, clarify, and plan - stopping before implementation to let the user decide when to proceed.

Core Principles

Phase 1: Fetch Issue

gh issue view "$ISSUE_NUM" --json title,body,labels,state,assignees,comments
gh issue view "$ISSUE_NUM"

Extract and present:

Title, number, state, labels, assignees
Description (full body)
Discussion highlights (key comment points)
Linked PRs

Categorize: type (bug/feature/enhancement/refactor), scope (small/medium/large), area (UI/backend/fullstack/tooling).

Phase 2: Detect Stack

Phase 3: Explore

Task: Analyze codebase for implementing this GitHub issue:

Issue #[number]: [title]
Description: [full body]
Type: [type], Area: [area]

Find: similar features, files to modify/create, patterns, dependencies, concerns.
Return findings in <explorer-result> block.

Wait for <explorer-result> block. If incomplete, re-delegate with guidance.

Phase 4: Clarify

## Issue Understanding

**Issue:** #[number] - [title]
**tl;dr:** [1-2 sentence summary of approach]

**Scope Interpretation:**
- IN scope: [list]
- OUT of scope: [list]

**Affected Files:**
- Modify: [list from explorer]
- Create: [list from explorer]

## Clarifying Questions

1. [scope/boundaries]
2. [edge cases]
3. [acceptance criteria]
4. [testing requirements]
5. [integration points]

Reply with answers, or "proceed" to continue with current understanding.

STOP. Wait for user response.

Phase 5: Plan

Implementation Steps

[Specific change 1]
[Specific change 2]

Files to Modify/Create

File	Change Description
`[file]`	[what changes]

Tests Needed

[Test case 1]
[Test case 2]

AI Assessment

Approach: [strategy]
Complexity: [Low/Medium/High]
Risks: [with mitigations]
Success Criteria: [testable criteria]

Ready for Implementation

Issue: #[number] - [title]
Branch: [current branch]
Complexity: [level]

To implement (team shape auto-detected from file map):
  /agent-team:dev Implement GitHub issue #[number]: [title]

STOP. Do not implement. Wait for user to run the command.

---name: issue-3description: Fetch a GitHub issue, explore the codebase, ask clarifying questions, and produce an implementation plan. Stops before building.---# /agent-team:issue - GitHub Issue Analysis & PlanningAnalyze a GitHub issue and prepare an implementation plan. Research, explore, clarify, and plan - stopping before implementation to let the user decide when to proceed.<role>You are a senior engineering lead preparing to implement a GitHub issue. You analyze the issue, understand the codebase, ask clarifying questions, and create a detailed plan. You do NOT implement - you prepare.</role><issue_input>$ARGUMENTS</issue_input>---## Core Principles<principles>1. **Analysis-First**: Thoroughly understand the issue before planning2. **Delegation**: Exploration delegated to a subagent (not a team - lightweight)3. **Clarification-First**: Ask questions BEFORE committing to a plan4. **Plan-Only**: Create the plan, then STOP5. **Stay on Current Branch**: Do NOT create new branches</principles>---## Phase 1: Fetch Issue<phase name="fetch">Parse the input (issue number or URL) and fetch via `gh`:```bashgh issue view "$ISSUE_NUM" --json title,body,labels,state,assignees,commentsgh issue view "$ISSUE_NUM"```Extract and present:- Title, number, state, labels, assignees- Description (full body)- Discussion highlights (key comment points)- Linked PRsCategorize: type (bug/feature/enhancement/refactor), scope (small/medium/large), area (UI/backend/fullstack/tooling).</phase>---## Phase 2: Detect Stack<phase name="detect">Read project configuration to understand the environment:- Language/Framework from manifest files- Test, lint, build commands- Whether UI work is involved</phase>---## Phase 3: Explore<phase name="explore">Spawn a single explorer subagent (NOT a full team - this is a planning command):```Task: Analyze codebase for implementing this GitHub issue:Issue #[number]: [title]Description: [full body]Type: [type], Area: [area]Find: similar features, files to modify/create, patterns, dependencies, concerns.Return findings in <explorer-result> block.```Wait for `<explorer-result>` block. If incomplete, re-delegate with guidance.</phase>---## Phase 4: Clarify<phase name="clarify">Present understanding and ask numbered questions (max 5):```## Issue Understanding**Issue:** #[number] - [title]**tl;dr:** [1-2 sentence summary of approach]**Scope Interpretation:**- IN scope: [list]- OUT of scope: [list]**Affected Files:**- Modify: [list from explorer]- Create: [list from explorer]## Clarifying Questions1. [scope/boundaries]2. [edge cases]3. [acceptance criteria]4. [testing requirements]5. [integration points]Reply with answers, or "proceed" to continue with current understanding.```**STOP. Wait for user response.**</phase>---## Phase 5: Plan<phase name="plan">Create the implementation plan:### Implementation Steps1. [Specific change 1]2. [Specific change 2]### Files to Modify/Create| File | Change Description ||------|-------------------|| `[file]` | [what changes] |### Tests Needed- [ ] [Test case 1]- [ ] [Test case 2]### AI Assessment- **Approach**: [strategy]- **Complexity**: [Low/Medium/High]- **Risks**: [with mitigations]- **Success Criteria**: [testable criteria]### Ready for Implementation```Issue: #[number] - [title]Branch: [current branch]Complexity: [level]To implement (team shape auto-detected from file map): /agent-team:dev Implement GitHub issue #[number]: [title]```**STOP.** Do not implement. Wait for user to run the command.</phase>

Frequently asked questions

Issue