[ult666666]

What changes

A voice and structure pass. No library code changes, and the README doctests are unchanged.

README (434 to 195 lines), organised as a library: install, one example, an API sketch, then how it works.

• The first screen says what quantkit does and how to install it, then shows the first doctest block. The "Note: educational software, not investment advice" box moved into the limits.
• The bold-lead-in feature list became an API sketch table of the five modules and the CLI.
• The other two doctest blocks stay in the README (CI runs them against the source tree and an installed wheel), with shorter notes after them.
• Five one-sentence limits replace five dense paragraphs. There was no AI note; there is now one plain line.

Before	After
"> Note: quantkit is educational and research software. It is not investment advice, and nothing in this repository is a recommendation to buy or sell anything."	First limit: "This is educational software, not investment advice, and there is no data downloader: bring a CSV of daily closes."
"Quantitative finance code fails quietly: a sign slip in a Greek... quantkit is a small, readable library built around verification."	"Each model is tested against something it did not produce itself: a textbook value, a closed form, an independent implementation or an identity that must hold."
"with 2.5 exceptions expected in 250 days, the nominal 5% Kupiec test rejects a correct model 9.5% of the time, and the two Christoffersen tests about 1%."	"with so few exceptions expected, a correct model fails the nominal 5% Kupiec test 9.5% of the time."
"Data and verification. There is deliberately no data downloader..."	"There is no data downloader: bring your own CSV with ISO dates." (docs/DESIGN.md)

Relocated, not removed.

• docs/CLI.md: every command-line transcript, unchanged.
• docs/EXAMPLES.md: the five figure and study write-ups, numbers unchanged, bold lead-ins turned into headings.
• docs/DESIGN.md: the per-component checks table, the data-flow diagram, the full limitations and roadmap, and the references. Its section "Questions worth being able to answer" is now "Common questions".

Verification

• uv run pytest --doctest-glob="README.md" --cov=quantkit --cov-fail-under=90: 435 passed, coverage 99.62%. ruff check, ruff format --check and mypy pass.
• The wheel step from CI, run locally: uv build --wheel, a fresh venv, then python -m doctest README.md from outside the repo. 34 examples passed, the same count as the old README. quantkit var-backtest --help works.
• Re-ran every command now in docs/CLI.md and the five example scripts (in an exported copy, so the figures in the repo were not rewritten). All output matches the text that moved.
• annualized_mean_returns on the README's synthetic prices gives −10.85% for CCC, matching the README's −10.9%.
• Relative links and anchors in the changed Markdown files resolve.

Limits

• docs/WALK_FORWARD.md is unchanged.

https://claude.ai/code/session_016h2RvhZFj77hPxFn8TQUqq

---

Generated by Claude Code