Skip to content
Back to skills

Park Idea

ASecurity

Quickly park ideas with optional metadata without breaking the current work flow.

  • 3 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added September 30, 2026
ai-agentsgobashapidatabaseperformancedocumentation

Works with

  • api

Security analysis

A100/100

Scanned September 30, 2026

npx -y skills add jskswamy/skills --skill park-idea --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Park Idea?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Park Idea
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/jskswamy-park-idea/badge)](https://www.skillsdirectory.com/skills/jskswamy-park-idea)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: park-idea
description: Quickly park ideas with optional metadata without breaking the current work flow.
---

# Park Idea Skill

Quickly park ideas with optional metadata. This command provides more control than the auto-invoked park-idea skill while maintaining the same "capture fast, return to work" philosophy.

## Arguments

| Argument | Short | Type | Default | Description |
|----------|-------|------|---------|-------------|
| `idea` | | string | (required) | The idea to park |
| `--task` | `-t` | string | (auto) | Link to specific task ID |
| `--priority` | `-p` | number | 4 | Priority override (0-4) |
| `--tags` | | string | `parked-idea` | Additional comma-separated tags |
| `--quick` | `-q` | boolean | false | Minimal output, no confirmation |

**Examples:**
```
/park "Add caching to the API"
/park "Consider rate limiting" -t abc123
/park -p 3 "Refactor auth module for testability"
/park --tags "performance,optimization" "Lazy load images"
/park -q "Remember to update docs"
```

---

## Execution Flow

### Step 1: Parse Arguments

Extract the idea text and all optional flags.

If no idea provided, prompt:
```
What idea would you like to park?
```

### Step 2: Gather Context (Automatic)

Silently gather context without prompting:

1. **Current task (if --task not provided):**
   ```bash
   bd list --status=in_progress --json 2>/dev/null | head -1
   ```

2. **From conversation context:**
   - What file/code was being discussed
   - What triggered this idea (recent topic)

### Step 3: Create or Update Parked Issue

**If creating a new issue:**
```bash
bd create "PARKED: {brief title from idea}" \
  -t task \
  -p {priority} \
  --status deferred \
  -l parked-idea \
  --description "## Idea
{user's idea text}

## Context
Parked while working on: {task-id or 'no active task'}
Triggered by: {what was being discussed}
File context: {current file if any, or 'N/A'}"
```

For additional tags, add multiple `-l` flags:
```bash
bd create ... -l parked-idea -l performance -l database
```

**If updating an existing related issue:**
When you find an existing issue that matches the parked idea, you MUST:
1. Add notes with the new context
2. Set status to deferred
3. Add the parked-idea label

```bash
bd update {existing-id} --status deferred --notes "{new context}"
bd label add {existing-id} parked-idea
```

This ensures the issue appears in `/parked list`.

### Step 4: Link to Source Task (if applicable)

If there's a current in-progress task or `--task` was provided:

```bash
bd dep add {new-idea-id} {task-id}
```

This creates a "discovered-from" relationship.

### Step 5: Confirm and Return

**If `--quick` is NOT set:**
```
Parked: {id} - PARKED: {title}
Priority: P{priority}
Linked to: {task-id or "none"}

Review parked ideas later: /parked list
```

**If `--quick` IS set:**
```
Parked as {id}!
```

Then immediately return focus to the previous work.

---

## Priority Guidelines

| Priority | When to Use |
|----------|-------------|
| P0-P1 | Rarely for parked ideas - these should probably be real tasks |
| P2 | Good ideas that should be done soon after current work |
| P3 | Solid ideas for later consideration |
| P4 | Default - backlog ideas to review eventually |

---

## Tag Patterns

Default tag: `parked-idea` (always included for filtering)

Common additional tags:
- `performance` - Performance optimization ideas
- `refactor` - Code quality improvements
- `feature` - New feature ideas
- `bug` - Potential bugs noticed
- `tech-debt` - Technical debt items
- `docs` - Documentation improvements

**Example with tags:**
```
/park --tags "performance,database" "Index the user_sessions table"
```

Creates issue with labels: `parked-idea,performance,database`

---

## Context Detection

The command automatically captures context:

| Context | How Detected |
|---------|--------------|
| Current task | `bd list --status=in_progress` |
| Current file | From conversation (what file was being discussed) |
| Trigger | Recent topic in conversation |

This context helps when reviewing parked ideas later - you'll remember why you had the idea.

---

## Error Handling

### No idea provided
```
Error: No idea provided.

Usage: /park <idea> [--flags]
Example: /park "Add caching to API"
```

### Invalid priority
```
Error: Invalid priority "{value}". Priority must be 0-4.
Using default priority P4 for parked ideas.
```

### Beads not initialized
```
Error: Beads not initialized in this project.

Run: bd init
Then try again.
```

---

## Delegation Notes

This command provides a direct interface to the park-idea skill with:
- Explicit argument parsing
- Priority override capability
- Custom tagging support
- Quick mode for minimal interruption

The core parking logic follows the same principles as the skill:
- Maximum 1 question (none if idea is clear)
- Auto-capture context
- Return to work fast

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.