Sim Trading MVP

Run a paper-trading / simulated investing workflow with explicit style selection, fixed risk rules, three decision windows per trading day, optional cron set...

This skill appears to do what it says: run a disciplined paper-trading loop, keep account and trade logs, and optionally schedule decision jobs. Before installing or enabling automation: 1) Confirm the filesystem paths (account.json, trades.jsonl) so the agent doesn't read/write unexpected locations. 2) Review and approve any cron commands the skill would add — don't allow automatic scheduling without your consent. 3) Keep your Finnhub (or other market-data) API key in your local runtime (.env or secret store); do not put it into the skill files or public repos. 4) Inspect the included script (update_account.py) — it's small and benign; run it manually first if you want. 5) If you don't want the agent to act autonomously on a schedule, avoid adding the cron jobs or restrict the skill's invocation. If you want more assurance, run the skill in an isolated session or workspace and verify the created files and cron entries match your expectations.

Type: OpenClaw Skill Name: sim-trading-mvp Version: 0.2.1 The skill implements a comprehensive paper-trading simulator with clear instructions for the AI agent to maintain a model portfolio. It includes a Python script (scripts/update_account.py) for calculating account equity and return percentages, and provides templates for cron-based automation (references/cron-setup.md) and reporting. The instructions in SKILL.md emphasize data integrity, explicitly forbid the fabrication of market data, and provide sound security advice regarding the handling of API keys (e.g., using .env files instead of hardcoding). All high-risk capabilities, such as file I/O and scheduled tasks, are directly aligned with the stated purpose of a persistent, automated trading simulation.