Best use case
new is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Create a GitHub issue (or local issue with --local)
Teams using new should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/new/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How new Compares
| Feature / Agent | new | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Create a GitHub issue (or local issue with --local)
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.
SKILL.md Source
# New Issue - Create GitHub Issue
Create a GitHub issue via `gh issue create`. **You do the work directly.**
Use `--local` for offline/private work in `.devloop/issues/`.
## Step 1: Parse Arguments and Check GitHub
Check `$ARGUMENTS` for title/description and `--local` flag.
```bash
"${CLAUDE_PLUGIN_ROOT}/scripts/check-gh-setup.sh"
```
**If `preferred_method` is "none":**
- `gh_installed: false`: Display install instructions, fall back to `--local`
- `gh_authenticated: false`: Prompt `gh auth login`, fall back to `--local`
**If "gh" or "curl":** Continue to GitHub Mode (unless `--local` specified).
---
## Step 2: GitHub Mode
### 2.1 Collect Info
```yaml
AskUserQuestion:
questions:
- question: "What type of issue?"
header: "Type"
multiSelect: false
options:
- label: "Bug"
description: "Something isn't working correctly"
- label: "Feature"
description: "New functionality or enhancement"
- label: "Task"
description: "General work item or chore"
- question: "Add labels?"
header: "Labels"
multiSelect: true
options:
- label: "bug"
description: "Bug fix"
- label: "enhancement"
description: "New feature or improvement"
- label: "devloop"
description: "Related to devloop plugin"
- label: "documentation"
description: "Documentation update"
```
Auto-add type label: Bug → "bug", Feature → "enhancement".
### 2.2 Title and Body
If no title in arguments, prompt for it. Then ask for description (what, why, context, acceptance criteria).
### 2.3 Create Issue
```bash
gh issue create \
--title "${title}" \
--body "${body}" \
--label "${labels_comma_separated}" \
--repo $(git remote get-url origin | sed 's/.*[:/]\([^/]*\/[^/]*\)\.git/\1/')
```
Parse issue URL and number from output.
### 2.4 Next Actions
```yaml
AskUserQuestion:
questions:
- question: "Issue #N created. What next?"
header: "Next"
multiSelect: false
options:
- label: "Start work"
description: "Create plan from this issue"
- label: "Create more"
description: "Create another issue"
- label: "View issues"
description: "List all open issues"
- label: "Done"
description: "Return to conversation"
```
Route: Start → `/devloop:plan --from-issue N`, More → loop, View → `/devloop:issues`, Done → exit.
---
## Step 3: Local Mode
**Only if `--local` specified or GitHub unavailable.**
### 3.1 Collect Info
```yaml
AskUserQuestion:
questions:
- question: "What type of issue?"
header: "Type"
multiSelect: false
options:
- label: "Bug"
description: "Something isn't working"
- label: "Feature"
description: "New functionality"
- label: "Task"
description: "General work item"
- label: "Spike"
description: "Research or investigation"
- question: "Priority?"
header: "Priority"
multiSelect: false
options:
- label: "High"
description: "Urgent"
- label: "Medium (Recommended)"
description: "Normal priority"
- label: "Low"
description: "Nice to have"
- question: "Labels?"
header: "Labels"
multiSelect: true
options:
- label: "devloop"
description: "Related to devloop"
- label: "commands"
description: "Related to commands"
- label: "agents"
description: "Related to agents"
- label: "ux"
description: "User experience"
```
### 3.2 Type-Specific Details
**Bug:** Severity (Critical/Major/Minor/Cosmetic)
**Feature/Task:** Size estimate (XS/S/M/L)
**Spike:** Depth (Quick/Standard/Deep)
### 3.3 Title and Description
If no title, prompt. Then ask for description.
### 3.4 Generate Issue ID
```bash
ls .devloop/issues/${TYPE}-*.md 2>/dev/null | wc -l
```
Format: `{TYPE}-{NUMBER}` (e.g., FEAT-003).
### 3.5 Create Issue File
Write to `.devloop/issues/${TYPE}-${NUMBER}.md`:
```markdown
---
id: ${TYPE}-${NUMBER}
type: ${type}
title: ${title}
status: open
priority: ${priority}
created: ${ISO_DATE}
labels: [${labels}]
---
# ${TYPE}-${NUMBER}: ${title}
## Description
${description}
## Acceptance Criteria
- [ ] ${auto_criteria}
## Technical Notes
<!-- Implementation notes -->
## Resolution
<!-- Filled when done -->
```
### 3.6 Update Index
Add to `.devloop/issues/index.md` (create if needed).
### 3.7 Next Actions
Same as GitHub mode but for local issues.
---
## Quick Mode
If `$ARGUMENTS` has 5+ words, enable quick mode:
1. Auto-detect type: "bug", "fix", "broken" → Bug; "add", "new", "feature" → Feature; else → Task
2. Extract title from arguments
**GitHub:** Prompt confirmation only, then create.
**Local:** Create immediately.
---
## Examples
```bash
/devloop:new # Interactive GitHub
/devloop:new Fix login button not responding # Quick mode GitHub
/devloop:new Add dark mode support # Quick mode GitHub
/devloop:new --local Private refactoring notes # Local issue
/devloop:new --local # Interactive local
```
## Migration
- **Default**: GitHub issues (was local)
- **Local**: Use `--local` flag
- **.devloop/issues/**: Still supported for `--local`Related Skills
Example Skill
Brief description of what this skill does and the domain expertise it provides.
vulnerability-patterns
Index of vulnerability detection pattern skills. Routes to core patterns (universal) and language-specific patterns for security scanning.
vuln-patterns-languages
Language-specific vulnerability detection patterns for JavaScript/TypeScript, Python, Go, Java, Ruby, and PHP. Provides regex patterns and grep commands for common security vulnerabilities.
vuln-patterns-core
Universal vulnerability detection patterns applicable across all programming languages. Includes hardcoded secrets, SQL/command injection, path traversal, and configuration file patterns.
scan
Run a security assessment using deterministic static analysis tools with LLM-powered triage
results
View the most recent security scan results without re-running the scan
remediation-library
Index of security remediation skills. Routes to specialized skills for injection, cryptography, authentication, and configuration vulnerabilities.
remediation-injection
Security fix patterns for injection vulnerabilities (SQL, Command, XSS). Provides language-specific code examples showing vulnerable and secure implementations.
remediation-crypto
Security fix patterns for cryptographic vulnerabilities (weak algorithms, insecure randomness, TLS issues). Provides language-specific secure implementations.
remediation-config
Security fix patterns for configuration and deployment vulnerabilities (path traversal, debug mode, security headers). Provides language-specific secure implementations.
remediation-auth
Security fix patterns for authentication and authorization vulnerabilities (credentials, JWT, deserialization, access control). Provides language-specific secure implementations.
fix
Fix or guide remediation for a specific security finding from the latest scan report