[JarryShaw]

Please follow the guide below

• You will be asked some questions, please read them carefully and answer honestly

• Put an x into all the boxes [ ] relevant to your pull request (like that [x])

• Use Preview tab to see how your pull request will actually look like

• Searched for similar pull requests

• Followed the coding style (make pylint, make mypy, make isort)

• make test passes, and a test case covers the change. Narrow runs only, listed below.

• Added a changelog entry under docs/source/changelog/ and regenerated CHANGELOG.md, if the change is user-visible: N/A — centralised in docs(changelog): shared 1.5.0 changelog — long-lived, merges last (#610, #616, #617, #618, #620) #657

What is the purpose of your pull request?

Tick the commit type your subject line carries.

• fix — corrects a defect
• feat — adds a feature
• perf — changes performance, not behaviour
• refactor — changes neither behaviour nor performance
• test — tests only
• docs — documentation only
• ci — workflows or build tooling
• release — bumps the version or rolls up a distribution
• chore — anything else

---

Description

Part of #1103, the whole-repository prose sweep, for the repo slice. It follows the #719 rules: prose only, accuracy first, no churn and no timed context.

Accuracy -- claims that no longer matched the tree:

• README: len(flows.trace) is 1 on main, not 3; show .trace.tcp, 2.
• Makefile, lint.yml: make vermin stops at its redirect when vermin
  exits 1; it does not succeed through the editor line.
• create-release.yml: version_check's snippets run after setup-python;
  unit-tests.yml: all four callers pass gate-only.
• dispatch.py: 47 entries in eight tables, two degraded; coverage.toml
  adds four settings, not three.
• python-versions.toml, the pyproject zip-safe note, the Makefile bench
  note, lint.yml's typing_extensions count, CONTRIBUTING's registry count.

Timed context: about 95 "now", "today", "currently", "yet", "this PR"
and comment-history phrases removed across the workflows, configs,
example generators, the benchmark and util/bump_version.py.

Comments and docstrings only: changed .py files keep an identical AST
with string expressions blanked; YAML/TOML load identically except for
shell comments in two run: blocks. tests/project passes.

Checks:

• An AST guard holds on every changed .py file.
• YAML waiver. Two hunks change shell comment lines (# lines) inside run: scripts in create-release.yml and cron-vendor.yml. With those lines stripped, both scripts are identical; declare -f of each confirms nothing executes differently. Every other YAML file loads equal.
• tests/project gives 930 passed and 1 skipped, and every test that reads the edited paths passes.
• A merged Sphinx build of all nine slices adds no new warnings.
• A Sonnet cross-review sent this back once. After the fix round, a re-check returned GOOD TO GO.

Not covered here; a round-2 PR follows: the files that were only grepped, the stale IPX text in examples/generators/dispatch.py, and the lint.yml header counts, which need re-measuring.

Refs #1103

[JarryShaw]

Verdict: GOOD TO GO at 5b06a3d70. A Sonnet cross-review checked it (author: Opus), after one fix round.

• First review: the YAML guard failed on two run: scalars.
• Waiver: I waived the guard for shell comment lines. The re-check proved the condition:
  • only # lines differ, and the scalars are identical once those lines are stripped;
  • the scripts contain no heredocs, and declare -f output is identical;
  • controls confirm the check catches a real change.
• Makefile: the self-contradicting clause is gone, and make vermin RUN= was re-measured.
• Guards: the AST, TOML and non-comment line checks all hold.
• Tests: tests/project gives 930 passed.