Code development

Natural invoke

• "Implement gradient descent in template_code_project src/"
• "Add a generic CSV validator in infrastructure/"
• "New analysis method with tests and logging"

Inputs to confirm

• Task — what to implement.
• Layer — infrastructure (generic, 60% cov) vs project (domain, 90% cov).
• Project — if layer is project, from docs/_generated/active_projects.md.

Workflow

1. Place logic correctly — computation in infrastructure/ or projects/<n>/src/; scripts only orchestrate I/O and plotting.

1. Implement — type hints on public APIs; get_logger(__name__); custom exceptions extending TemplateError; Google docstrings.

1. Tests — real data, temp files, subprocess, or pytest-httpserver; no mocks (docs/rules/testing_standards.md).

1. Docs — update module/folder AGENTS.md + README.md when adding public surface.

1. Verify — ruff/mypy on touched paths; pytest with coverage gate.

Deliverables

• Module(s) with tests meeting coverage floor
• List of files touched and verification command output

Verification commands

bash
uv run python -m infrastructure.project.public_scope lint-paths | xargs uv run ruff check
uv run python -m infrastructure.project.public_scope source-paths | xargs uv run mypy
uv run pytest projects/<project>/tests/ --cov=projects/<project>/src --cov-fail-under=90 -q
# or infra:
uv run pytest tests/infra_tests/ --cov=infrastructure --cov-fail-under=60 -q

When NOT to use

• End-to-end feature spanning manuscript + pipeline → feature-addition
• Tests only for existing code → test-creation
• New infrastructure package skeleton → infrastructure-module

References

• docs/rules/ — normative standards
• docs/core/architecture.md — two layers, thin orchestrator
• Code patterns: references/patterns.md