| name | run_screener |
| description | Launch the Investment Screener suite (backend + frontend). Trigger when user says "run the screener", "start the app", "launch investment toolkit", "start servers", or "run investment screener".
|
| allowed-tools | Bash, Read, Write |
Run Investment Screener Skill
Identity
You are a utility agent focused on launching and monitoring the Investment Toolkit suite.
Purpose
Orchestrates the startup of the backend and frontend services via the unified python runner, ensuring a smooth user experience by only reporting final status once everything is ready.
Trigger Phrases
- "run the screener"
- "start the app"
- "launch investment toolkit"
- "start servers"
- "run investment screener"
Steps
- Launch: Execute the unified startup script from the root:
python3 run_investment_toolkit.py
- Monitor Prerequisites: Observe the logs as it installs Python and Node dependencies and builds the backend. Do NOT provide any URLs to the user during this phase, as they will not be active yet.
- Validate Startup: Wait until you see the "✅ Services Running!" message and confirmation that the Backend (3001) and Frontend (5173) are listening.
- Final Report: Once (and only once) the services are fully operational, provide the local URLs to the user and explain how to stop them (Ctrl+C).
Common Failures
- Port Conflict: If 3001 or 5173 are occupied. The script should handle clearing them, but if it fails, advise the user to check active processes.
- Node/Python missing: Ensure both runtimes are available in the shell path.
Data Architecture Note
Startup today only initializes JSON data files (see the onboarding guide's Phase 2). A
SQLite-backed domain data model is in active design (../../references/data-architecture/domain-data-model.md)
but not yet implemented — this skill does not run any schema initialization. If/when that model
is implemented, a schema-init step belongs in the Launch step above, not before.
