idea128-investlearn-studio-verifiable/AGENTS.md

AGENTS.md

Repository purpose

This repo hosts the InvestLearn Studio prototype: a verifiable, offline-first investment education & practice engine. The code here is intentionally small but structured to be extended by additional agents in the swarm.

Architecture & Tech Stack

• Language: Python 3.8+
• Packaging: pyproject.toml (setuptools backend)
• Key packages: pandas (analytics), pytest (tests)
• Source layout: src/idea128_investlearn_studio_verifiable

Key components

• dsl.py: simple DSL parser for learner objectives
• modules/: learning modules (risk_literacy, portfolio)
• simulation/: market simulation engine
• sync/: lightweight cryptographic ledger for attestations
• analytics/: toy analytics harness using pandas

Testing & Commands

Run the full verification and packaging pipeline locally:

./test.sh

What agents must follow

1. Read this file before changing packaging or tests. 2. Do not remove test.sh or change its behavior without updating the CI maintainers. 3. When adding dependencies, update pyproject.toml and ensure tests still run quickly. 4. Keep changes minimal and localized. 5. If you're adding new data files, update MANIFEST or package_data accordingly.