[noahgift]

On a build without trueno-explain, only about was overridden for ptx, so apr ptx --help (the long form) still came from the doc comment and lost the marker. long_about now carries the same text. A test reads clap's own Command for ptx and asserts the marker is present in both forms exactly when the feature is off. It failed before the fix and passes after (run on intel). The feature-on arm of the test is reasoned, not run.

3 files, +46/-1, no ci.yml. C314 bank: receipts/c314-4474-b33609df84-r4-bank.md (GRANT 3/3, manual argon lane).

Refs #4474, Refs #5328
keep-open: #4474 and #5328 stay open (umbrella and sibling bullets remain); a lane never closes issues.

🤖 Generated with Claude Code

[noahgift]

seat 1, aprender-seat-s11 @b33609df84 APPROVE
Diff read (3 files, +46/-1; no workflow/gate files). Fix: with trueno-explain off, the ptx variant overrode only about, so --help (long form) fell back to the doc comment and lost the marker; long_about now carries the same text, cfg-gated by the existing not(feature = "trueno-explain") attribute. The test (new ptx_help_marker_tests.rs, wired in commands_enum.rs like code_output_format_help_tests) reads clap's own Command, checks marker presence in both about and long_about equals !trueno-explain, on a 16 MiB stack as the sibling test does; it is two-armed so it does not pass on a constant. Not built or run by me; the author states red-before/green-after on intel and that the feature-on arm is reasoned, not run - I agree it follows from the cfg gate, as the doc comment carries no marker text. Changes only help text of an unavailable-feature stub.

[noahgift]

seat 2, aprender-seat-s12 (claude-sonnet-5-5) @b33609df84 APPROVE

Read the 3-file diff (+46/-1; no workflow or gate files). long_about on the no-trueno-explain ptx variant now carries the same marker text as about, so clap no longer derives the long form from the doc comment. New test ptx_help_marker_tests reads clap's own Cli::command() for ptx (16 MiB stack thread, as code_output_format_help_tests.rs does) and asserts marker presence == !trueno-explain for both about and long_about (falling back to about when no long form exists).

• Run on an archive of the head, fresh target dir, cargo test -p apr-cli --lib ptx_help: default features (trueno-explain off) 1 passed; with --features trueno-explain 1 passed. So the feature-on arm, which the PR body says was only reasoned, passes when run.
• Mutant: removing the long_about = ... line in my scratch copy makes the test fail (panic at the assert_eq!, rc 101), so the test discriminates the fix.
• Not run: no workflow; the C314 bank was not re-run by me.

[github-actions]

§13.11 rung 1 — quorum shadow verdict

S13-SHADOW pr=5456 head=92340b94184f01e4563ccbcfe5f8b30286166efa verdict=REFUSE class=Q1 arm_rc=1

Shadow mode: this records a verdict and merges nothing. A refusal
to arm is not a block (§13 adds zero rows to §7) — the pull request is
exactly as green as it was.

[noahgift]

seat 1, aprender-seat-s11 @92340b9418 APPROVE
Re-seat of my @b33609df84 approval. Delta is 2 lines of docs/audits/surface_audit.csv (git diff b33609d 92340b9 --stat: 1 file, +2/-2): apr ptx evidence_path extended_commands.rs:387 -> :407 and apr tune :409 -> :429. Verified at the head: line 407 of crates/apr-cli/src/extended_commands.rs is Ptx { and line 429 is Tune {, i.e. the variant lines, shifted by the added long_about marker. No Rust or workflow file changed. Earlier review stands.

[noahgift]

seat 2, aprender-seat-s12 (claude-sonnet-5-5) @92340b9418 APPROVE

Delta vs my approved b33609d, read in full: git diff --stat is one file, docs/audits/surface_audit.csv, 2 lines. No Rust, workflow or ci/ change.

• apr ptx evidence_path extended_commands.rs:387 -> :407; apr tune :409 -> :429 (both +20).
• Checked at the new head: line 407 is Ptx { and line 429 is Tune {, i.e. each now lands on its own command's enum variant in crates/apr-cli/src/extended_commands.rs, so the pointers follow the shifted file.
  No cargo run needed; nothing else changed.