IDEA Storer (IS) Integration
A CLI tool for capturing development ideas that AI agents can understand.
Discovery
Check for ideas in this order:
- •Project-specific:
.ideas.jsonin current directory or git root - •Global fallback:
~/.config/ideastorer/ideas.json
Reading Ideas
When user says "check IS", "my ideas", or "what should I work on":
- •Read the
.ideas.jsonfile using the Read tool - •Parse the JSON and extract active ideas
- •Present relevant ideas based on current context
- •Suggest which ideas relate to current work
JSON Schema
json
{
"version": "1.0.0",
"ideas": [
{
"id": "abc12345",
"content": "The idea text",
"tags": ["tag1", "tag2"],
"priority": "low|medium|high",
"status": "active|done|archived",
"created_at": "ISO8601",
"context": {
"project": "project-name",
"git_branch": "branch-name"
}
}
]
}
CLI Commands
bash
is "quick idea" # Add idea instantly is add "idea" -t tag # Add with tags is list # Show recent ideas is list -t tag # Filter by tag is search "query" # Search ideas is done <id> # Mark completed is delete <id> # Remove idea is export # Export as JSON is export --format md # Export as Markdown
Workflow Integration
Before Starting Work
code
"Check IS for ideas related to [current task]" → Read .ideas.json → Filter by relevant tags → Present actionable items
During Implementation
- •Reference idea IDs in commit messages
- •Cross-check implementation against stored ideas
- •Suggest related ideas that could be addressed
After Completion
code
"Mark idea <id> as done" → Run: is done <id> → Confirm completion
Example Interactions
User: "What should I work on?"
Action: Read .ideas.json, present high-priority active ideas
User: "Check IS for auth ideas" Action: Search ideas with "auth" or tag "authentication"
User: "Add to IS: implement caching"
Action: Run is add "implement caching" -t performance