Requirements for an Autonomous Retail Order Engine Running Unattended for a Decade

A private research project produced three frozen, low-turnover trading strategies for a retail account, a nightly signal pipeline, a data-freshness guard and a dry-run order planner. The component that would actually place orders, an order engine that runs unattended for ten years on a home machine, does not exist yet. Before the project is archived, this paper records what that engine must do and, for the parts that do exist, how they work in enough detail to rebuild them. It is a requirements and design paper, not a report of a built system. Every requirement is derived from a concrete artefact, rule, bug or documented decision of the project, and the appendix lists 122 of them with ID, rationale and provenance, marking each as existing (E), partially existing (P) or new (N). The paper specifies the planner (bands, sizing, limit prices, trimming), the reference price and execution-day rules, the retail fee schedule and why it forbids small orders, the first-deployment case, liquidity checks, freshness gates and their parameters, data rules learned from bugs (a persisted rate-limit lock, completed sessions only, a daylight-saving mismatch), the instrument map with a wrong-ISIN bug, idempotency, an order state machine with timeouts, limits and kill switches, a staged human-in-the-loop policy, credentials, scheduling and a verification plan. No part of the real-order path has been built or tested; every threshold that the project did not fix is marked as open or as a suggestion.Read next: The strategy the engine would run comes from the pre-registered out-of-sample test. Its signals are produced with leak-resistant point-in-time backtesting. Operational lessons are shared with running an autonomous AI agent on trading research.This preprint is part of an eight-paper series on retail equity research from public data by Christopher Helm, covering German and Nordic small caps, backtesting methodology, data sources and AI-assisted research. All papers in the series:How Wise Is the Crowd? Timing and Selection Skill of Retail Social-Trading PortfoliosA Leak-Resistant Point-in-Time Backtesting Protocol and Engine for Retail Equity StrategiesThree Pre-Registered Bets: A One-Shot Out-of-Sample Test of Retail Strategy CandidatesWhat Did Not Work: Negative Results on Event-Driven Signals in German and Nordic Small CapsFree, Point-in-Time Public Data for German and Nordic Small-Cap Research: Sources, Coverage and PitfallsRunning an Autonomous AI Agent on Quantitative Trading Research: Protocol, Supervision and Failure ModesFrom Pre-Announcement to Verified Facts: Design of a Primary-Source Report-Analysis Funnel for Small-Cap EquitiesChristopher Helm is an entrepreneur, consultant, and researcher. More research and articles on his blog.