issue
Create standalone work items (TASK, BUG, or SPIKE) with AI-assisted type detection. Use when you need to track work that needs doing, exploration needed, or something broken.
About this skill
This AI agent skill, `/issue`, automates the creation of structured work items within a project's local file system. It intelligently detects the type of work—TASK, BUG, or SPIKE—based on a natural language description provided by the user. The skill guides the user through the process, prompting for missing information like the project name and offering to link the new issue to existing specification documents. Users benefit by streamlining their issue creation workflow, ensuring consistency in how work items are categorized and tracked. It's ideal for developers and project managers who need a quick yet structured way to log new development tasks, identify and track bugs, or document time-boxed research efforts. The skill integrates seamlessly with a local `ideas/[project]/issues/` directory, creating a foundation for further planning and progress tracking.
Best use case
The primary use case for this skill is to efficiently create and categorize new project issues (tasks, bugs, or spikes) using natural language, directly within a structured file system. It benefits developers, project managers, and teams looking to quickly document and organize work items, ensuring consistent tracking and easy reference to project specifications.
Create standalone work items (TASK, BUG, or SPIKE) with AI-assisted type detection. Use when you need to track work that needs doing, exploration needed, or something broken.
A new markdown file (e.g., `TASK.md`, `BUG.md`, `SPIKE.md`) will be created in the `ideas/[project]/issues/` directory, containing the issue details and type.
Practical example
Example input
/issue "Fix the broken pagination on the user dashboard"
Example output
This sounds like a **BUG**. Create as issue 005 for 'my-project'? (yes / task / spike) > yes Created: ideas/my-project/issues/005-fix-pagination/BUG.md
When to use this skill
- When you need to create a new task, bug, or spike for a project.
- When you want AI to intelligently suggest the type of work item based on your description.
- When managing work items across different projects and linking them to specific specs.
- When you need a structured, file-based system for tracking project issues and progress.
When not to use this skill
- When you need to modify or close an existing issue (this skill is for creation).
- When your project exclusively uses an external issue tracker (e.g., Jira, GitHub Issues) and you don't want local duplication.
- When you need to perform complex project planning, resource allocation, or generate detailed reports beyond creating individual work items.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/issue/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How issue Compares
| Feature / Agent | issue | Standard Approach |
|---|---|---|
| Platform Support | Claude | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | easy | N/A |
Frequently Asked Questions
What does this skill do?
Create standalone work items (TASK, BUG, or SPIKE) with AI-assisted type detection. Use when you need to track work that needs doing, exploration needed, or something broken.
Which AI agents support this skill?
This skill is designed for Claude.
How difficult is it to install?
The installation complexity is rated as easy. You can find the installation instructions above.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
Related Guides
Best AI Skills for Claude
Explore the best AI skills for Claude and Claude Code across coding, research, workflow automation, documentation, and agent operations.
ChatGPT vs Claude for Agent Skills
Compare ChatGPT and Claude for AI agent skills across coding, writing, research, and reusable workflow execution.
Top AI Agents for Productivity
See the top AI agent skills for productivity, workflow automation, operational systems, documentation, and everyday task execution.
SKILL.md Source
# /issue
Create standalone work items through natural conversation with AI-assisted type detection.
## Usage
```bash
/issue # Start conversation
/issue "Implement authentication" # AI detects: TASK
/issue "Compare GraphQL vs REST" # AI detects: SPIKE
/issue "Broken link in project-brief" # AI detects: BUG
/issue --project coordinatr # Create for specific project
```
## Issue Types
| Type | Purpose | Detection Keywords |
|------|---------|-------------------|
| **TASK** | Work that needs doing | implement, create, add, build, write, set up |
| **SPIKE** | Time-boxed exploration | compare, vs, should we, evaluate, explore, research, feasibility |
| **BUG** | Something broken/wrong | fix, broken, incorrect, outdated, wrong, error |
## File Structure
```
ideas/[project]/issues/
└── 001-implement-auth/
├── TASK.md # or SPIKE.md or BUG.md
├── PLAN.md # Created by /plan
└── WORKLOG.md # Progress tracking
```
## Execution Flow
### 1. Gather Context
If no description: "What needs to be done?"
If no project: "Which project?"
### 2. Detect Type
Analyze description keywords, present detection:
> "This sounds like a **TASK**. Create as issue 002? (yes / spike / bug)"
### 3. Determine Next Issue Number
```bash
ls ideas/[project]/issues/ | grep -E '^[0-9]{3}-' | sort -n | tail -1
```
### 4. Ask About Spec Section
Check if project has a spec:
```bash
Glob: spaces/[project]/docs/specs/*.md
```
If spec exists:
> "Which spec section does this implement?"
> - docs/specs/required-features.md#authentication
> - docs/specs/required-features.md#documents
> - none (standalone task)
### 5. Create Issue Files
**TASK.md:**
```markdown
---
status: open
created: YYYY-MM-DD
implements: docs/specs/required-features.md#authentication # or empty if standalone
depends_on: []
---
# TASK-###: [Title]
## Description
[What needs to be done and why]
## Implements
**Spec Section:** [docs/specs/required-features.md#authentication](../../spaces/[project]/docs/specs/required-features.md#authentication)
**Requirements from spec:**
- [Requirement 1 from spec]
- [Requirement 2 from spec]
## Acceptance Criteria
- [ ] Criterion 1
- [ ] Criterion 2
## Context
[Background, relevant decisions, technical notes]
```
**SPIKE.md:**
```markdown
---
status: open
created: YYYY-MM-DD
timebox: X hours
---
# SPIKE-###: [Title]
## Questions
- Question 1?
- Question 2?
## Approaches to Explore
1. Approach A
2. Approach B
## Findings
(filled after exploration)
## Recommendation
(filled after exploration)
```
**BUG.md:**
```markdown
---
status: open
created: YYYY-MM-DD
---
# BUG-###: [Title]
## What's Broken
[Description of the problem]
## Location
[File paths, URLs, etc.]
## Expected Behavior
[What should happen]
## Steps to Reproduce
1. Step 1
2. Step 2
```
### 6. Mark Spec Section In Progress (if implements spec)
If the issue implements a spec section, update the inline status markers in the spec:
```markdown
# Before
- ⏳ User registration with email/password
# After
- 🚧 User registration with email/password
```
The `/complete` command will mark these ✅ when done.
### 7. Next Steps
- For TASK/BUG: Suggest `/plan ###` to create implementation phases
- For SPIKE: Suggest `/plan ###` to create exploration plan
## Status Values
| Status | Meaning |
|--------|---------|
| `open` | Not started |
| `in_progress` | Being worked on |
| `blocked` | Waiting on something |
| `complete` | Done |
## Spec Integration
### Task Scoping
**One TASK = One requirement line item**
A TASK should be atomic and shippable:
- Implements exactly one spec requirement
- Can be pushed to main independently
- Updates one `⏳` → `✅` marker when complete
```markdown
# Spec line items (each becomes a TASK)
- ⏳ User registration with email/password ← TASK-002
- ⏳ User login with JWT token ← TASK-003
- ⏳ Password reset flow ← TASK-004
```
**Don't create:** "TASK: Implement Authentication" (too broad)
**Do create:** "TASK: User registration endpoint" (one requirement)
### implements: Field
```yaml
# Points to the specific requirement
implements: docs/specs/required-features.md#user-registration-with-email-password
```
This creates a direct link between work items and the exact requirement they fulfill.
## Workflow
```
/spec → /issue → /plan → (work) → /complete
↓
implements: spec section
```Related Skills
monday-automation
Automate Monday.com work management including boards, items, columns, groups, subitems, and updates via Rube MCP (Composio). Always search tools first for current schemas.
linear-claude-skill
Manage Linear issues, projects, and teams
linear-automation
Automate Linear tasks via Rube MCP (Composio): issues, projects, cycles, teams, labels. Always search tools first for current schemas.
jira-automation
Automate Jira tasks via Rube MCP (Composio): issues, projects, sprints, boards, comments, users. Always search tools first for current schemas.
executing-plans
Use when you have a written implementation plan to execute in a separate session with review checkpoints
clickup-automation
Automate ClickUp project management including tasks, spaces, folders, lists, comments, and team operations via Rube MCP (Composio). Always search tools first for current schemas.
build
build
basecamp-automation
Automate Basecamp project management, to-dos, messages, people, and to-do list organization via Rube MCP (Composio). Always search tools first for current schemas.
issues
Interact with GitHub issues - create, list, and view issues.
create-issue-gate
Use when starting a new implementation task and an issue must be created with strict acceptance criteria gating before execution.
Beads Issue Tracking Skill
> **Attribution**: [Beads](https://github.com/steveyegge/beads) by [Steve Yegge](https://github.com/steveyegge)
workspace-surface-audit
Audit the active repo, MCP servers, plugins, connectors, env surfaces, and harness setup, then recommend the highest-value ECC-native skills, hooks, agents, and operator workflows. Use when the user wants help setting up Claude Code or understanding what capabilities are actually available in their environment.