Hermes Atlas
Automations & use cases

Virtual Investor

balsimpson/virtual-investor

Harper, a long-only virtual portfolio investing skill for Hermes Agent with an auditable SQLite ledger

In short

Virtual Investor is a Hermes Agent skill, named Harper, that researches markets and manages a simulated long-only portfolio without ever placing real orders.

What Virtual Investor does

Harper is a long-only virtual portfolio engine and conversational investing skill. It combines sourced market research with deterministic risk checks, virtual execution, marked-to-market accounting, forecast calibration and an auditable SQLite ledger. Decisions are evidence-backed BUY, SELL or NO_TRADE calls, and intraday or position-style theses carry explicit invalidation rules. Versioned market adapters let it work across markets, with transparent fallbacks for newly configured ones.

It never places real orders, connects to a broker, borrows to trade or opens short positions. The runtime is Python standard library with local SQLite storage, by default at ~/.hermes/data/virtual-investor/portfolio.db. After install, you start a new Hermes conversation and invoke /harper, which asks for one onboarding detail at a time, including virtual starting cash. Automation is optional: Harper offers it only after full research access is verified, asks for confirmation before enabling Hermes jobs, and can send updates to a messaging destination you choose or keep reports local. An optional Convex dashboard sync is available.

Key features

  • Evidence-backed BUY, SELL and NO_TRADE decisions
  • Long-only risk controls enforced by the portfolio engine
  • Auditable local SQLite ledger with marked-to-market accounting
  • Candidate screening, research memory and forecast scoring
  • Versioned market adapters for different markets
  • Optional scheduled Hermes automation and Convex dashboard sync

When to use it

  • Paper-trading ideas with a Hermes agent that explains each decision
  • Getting scheduled portfolio reports in a messaging app you already use
  • Auditing how a simulated portfolio decision was made from the ledger

Who it is for: Hermes Agent users who want a simulated investing assistant for research and paper trading.

How it fits with Hermes Agent

Built for Hermes Agent as an installable skill, using Hermes web search and extract tools for research and Hermes jobs for optional automation.

How to install Virtual Investor

These commands are copied from the project's README. Check the repository for the latest steps before you run them.

hermes skills install balsimpson/virtual-investor/skills/harper

Requirements: Python 3, and Hermes Agent with Web Search & Extract configured for full research use

Note: Harper is a simulation only: it tracks a virtual portfolio and never connects to a broker or places real orders.

FAQ

What is Virtual Investor?

Virtual Investor is a Hermes Agent skill named Harper that runs a long-only virtual portfolio. It researches markets, applies risk checks and records every virtual trade in a local SQLite ledger.

Does Virtual Investor place real trades?

No. Harper never places real orders, connects to a broker, borrows to trade or opens short positions. It only simulates a portfolio.

How do I install Virtual Investor?

Run hermes skills install balsimpson/virtual-investor/skills/harper, then start a new Hermes conversation or run /reset and invoke /harper to begin onboarding.

Similar automations for Hermes Agent

All automations

Related guides: Build agent teams with the Hermes kanban board · How to run Hermes Agent securely