Set up Cursor Cloud dev environment

[Mukku27]

Summary

Sets up the development environment for this AI-powered Inventory Management (Streamlit + SQLite + Gemini) app, verifies it works end-to-end, and gets CI green.

• Adds AGENTS.md with ## Cursor Cloud specific instructions capturing non-obvious startup caveats (no python3.11 binary, DB is gitignored and seeding is opt-in via --seed, app runs fully without API keys, pandasai intentionally absent).
• Bumps 8 pinned dependencies in constraints.txt to patched versions to clear pip-audit findings (cryptography 48.0.1, gitpython 3.1.50, idna 3.15, msgpack 1.2.1, pillow 12.2.0, pytest 9.0.3, tornado 6.5.7, urllib3 2.7.0); the regenerated lock also includes previously-omitted pip-tools subdeps.

No application code was modified.

CI

Was failing on pip-audit (24 vulns across 8 transitively-pinned packages from newly-published CVEs). After bumping the pinned versions, all 5 checks pass.

Environment verification

Installed dependencies into .venv (Python 3.12; project supports >=3.11), initialized + seeded the SQLite database, and ran the app.

Check	Command	Result
Lint	ruff check .	✅ All checks passed
Tests	python -m pytest	✅ 65 passed
Security	pip-audit	✅ No known vulnerabilities
Run	streamlit run app.py	✅ Serves on :8501 (health 200)

Hello-world walkthrough

Loaded the dashboard (500 seeded products, ~$123.9M inventory value), entered the natural-language query "How many products are in inventory?", and the app generated + executed SELECT COUNT(*) AS product_count FROM PRODUCT, returning 500.

inventory_app_nl_query_demo.mp4

Inventory dashboard
Natural language query result

_{To show artifacts inline, enable in settings.}

  

[cursor]

Bugbot is not enabled for your account, so this pull request was not reviewed.

Enable Bugbot in the Cursor dashboard to get automatic reviews on future PRs.