[teamleaderleo]

Base-red fix: the Rust panic ratchet step of cmux-tui.yml is red on feat-cmux-next for acpmux: clippy::string_slice 98 > baseline 94 (since 7acff13) and clippy::indexing_slicing above 693 (700 at e863077, 703 by 7f97b84). Every new acpmux hit in production code is rewritten so it can't panic and behaves the same.

Where the hits came from (blame), and what replaced them

commit	file	hits	now
84705bd (#26984)	hub/restart_resume.rs	4 string_slice, 1 indexing_slicing	char_prefix helper with str::get; get(..i); BoundedJsonWriter copies with take(remaining)
72c0200 (#28046)	cli/ls_table.rs, cli/run/ls.rs	6 indexing_slicing	get / get_mut in table() and render(); JsonSet::set for the ls params and sessions
95243d1 (#27685)	hub/cancel.rs	1	set("restarted", ..)
e863077 (#27809)	hub/remote_rows.rs	4	mark_peer_stale reads status, stale and lastSeen with get (a missing field compares the same as the Null that indexing returned)
9d318f9 (#28000)	cli/resolve.rs, hub/key_route.rs	4	choice_commands uses enumerate + checked_sub / get; key_route reads name and status with get
65b5de8 (#28016)	cli/orchestrate.rs	1	set(HARNESS_FROM_FIELD, ..)
7050ae3 (#27038)	hub/message_events.rs, hub/mod.rs, prompt_provenance.rs, server/outbox.rs, server/peer_forward.rs	7	writes via set; outbox reads via pointer (missing reads as Null, as before)
e0f3350 (#27614)	hub/paging.rs	2	the moot slot is read and cleared with get / get_mut
4a0b8f0 (#28172)	server/requests/briefs.rs	1	set("peersFailed", ..)
0a088aa (#28154)	hub/turns.rs	2	the stop reason and the stopped answer are read with get

JsonSet::set (cmux-json) inserts on an object, turns Null into an empty object first (as IndexMut does), and leaves any other value alone. IndexMut would have panicked on those. Every value written here is an object it just built.

Baseline

acpmux measured at 339e742 (fast-checks run 41057101016): indexing_slicing 688, string_slice 94, unwrap_used 143, all at or below the baselines (693 / 94 / 144). The baseline file is unchanged. New acpmux hits landed on base three times while this PR was open, so a baseline pinned to today's exact count would turn base red as soon as another one lands before this merges. A follow-up can lower it once base is quiet.

Tests

• restart_resume::multibyte_text_is_cut_on_char_boundaries: multibyte text through each of the four string cuts. It includes a 4096-byte scan cut inside a 3-byte char, a truncated tail cut at U+3000, and CJK text before PASSWORD=.
• ls_table_tests::a_short_or_empty_row_pads_without_panicking: an empty table, rows shorter than the header (they panicked before), and a row longer than the header.

Observed / Not run

• At 339e742 (run 41057101016), gated: cargo clippy --locked --keep-going -p acpmux -p cmux-tui --all-targets -- -D warnings passes (365994d had three unused import: cmux_json::JsonSet errors from this PR, fixed in the last commit). acpmux counts are 688 / 94 / 143, and cargo test -p acpmux --lib 2184 passed, 0 failed.
• The ratchet step is still RED at that run for a different crate, cmux-tui: clippy::indexing_slicing 861 > baseline 833, which this PR does not touch. A separate PR is fixing it.
• Earlier, at a0bd7c6: the ratchet step as written in the workflow, plus 26 restart_resume tests, all green (run 41023303857).
• Not run: macOS builds (no Swift change).

Review (gpt-6.1-sol, high, fast): SHIP at 339e742 (the import fix), and SHIP at 365994d for everything before it, with no findings (earlier SHIPs: a0bd7c6, 08fb305, 84671eb).

Changelog

none

risk:low (panic-free rewrites with the same output, tests, and a lower baseline).

Agent session: agent session

🤖 Generated with Claude Code

[github-actions]

cmux status _{(one comment, edited in place; each line is updated by the job that owns it)}

Review 339e742: HQ feed card · gallery before/after runs · checks

App of this head (app.zip, kept 1 day)

gh workflow run reload-build.yml -R manaflow-ai/cmux --ref feat-cmux-next -f tag=pr28134 -f ref=lane/acpmux-string-slice-ratchet -f platform=macos -f backend_mode=local

red: cmux-next checks (god files, concurrency, crash safety, l10n); NEW: 3 commit(s) change the sidebar's drawing code without a Sidebar-Look: trailer or an entry in plans/cmux-next/sidebar-owner.md is not red on the base 6607ecd; do not merge until green; see lane-rules

[teamleaderleo]

Review (the owner lane's independent review, at its land request): SHIP at a0bd7c6

Recorded by lane-gh-broker from this PR's owner lane, which asked to land this exact head after its independent review. next-autoland lands it once every red is the base's; a BLOCK at this head, a HOLD line or a hold label still stops it.

[blacksmith-sh]

Found 6 test failures on Blacksmith runners:

Failures

Test	View Logs
main.DeveloperIdAppTests.test_download_prints_the_active_profile_only/ test_download_prints_the_active_profile_only	View Logs
stubs: /System/Library/Frameworks/Security.framework/Versions/A/Security has no stub/ Security has no stub	View Logs
stubs: CoreFoundation.tbd: install-name /System/Library/Frameworks/ CoreFoundation.framework/Versions/A/CoreFoundation, missing 1: ['_CFRetain']/ CoreFoundation, missing 1: ['_CFRetain']	View Logs
stubs: CoreFoundation.tbd: install-name /System/Library/Frameworks/ CoreFoundation.framework/Versions/A/CoreFoundation, missing 1: ['_log2']/ CoreFoundation, missing 1: ['_log2']	View Logs
test_every_listed_hook_is_one_glaeda_released_with_the_check (main.SigningHooksInSt epWithGlaedaTests.test_every_listed_hook_is_one_glaeda_released_with_the_check)/ test_every_listed_hook_is_one_glaeda_released_with_the_check (main.SigningHooksInSt epWithGlaedaTests.test_every_listed_hook_is_one_glaeda_released_with_the_check)	View Logs
test_the_repository_highlights_validate (main.DigestTests.test_the_repository_highl ights_validate)/ test_the_repository_highlights_validate (main.DigestTests.test_the_repository_highl ights_validate)	View Logs

^{Need help on this PR? Tag @codesmith-bot with what you need.}

[teamleaderleo]

Base merge 0185b3d (clean; the PR diff is unchanged from a0bd7c6, and base touched neither restart_resume.rs nor the panic baseline), so the gpt-6.1-sol SHIP at a0bd7c6 carries. This also reruns 'cmux-tui artifacts / publish to R2', which failed only on the attestation step's API rate limit (infra, not this diff).

[teamleaderleo]

Review (the owner lane's independent review, at its land request): SHIP at 365994d

Recorded by lane-gh-broker from this PR's owner lane, which asked to land this exact head after its independent review. next-autoland lands it once every red is the base's; a BLOCK at this head, a HOLD line or a hold label still stops it.

[teamleaderleo]

Review (the owner lane's independent review, at its land request): SHIP at 339e742

Recorded by lane-gh-broker from this PR's owner lane, which asked to land this exact head after its independent review. next-autoland lands it once every red is the base's; a BLOCK at this head, a HOLD line or a hold label still stops it.

[teamleaderleo]

Red checks at 339e742, classified:

• clippy (changed cmux-tui packages): green now. At 365994d this PR caused it (3 unused JsonSet imports), and 339e742 fixed it.
• protocol contract / inventory › "Check public resource API boundary": base red. All 5 diagnostics are in crates/cmux-tui/src/cli.rs:767-771 (list-surfaces, new-surface and the others). The cmux-tui crate in this PR is byte-identical to origin/feat-cmux-next, and running the script locally on this head gives the same 5 lines.
• cmux-next checks (String tables) and release-notary: base reds this PR does not touch (no Swift, l10n or release-notes changes).

[github-actions]

Shipped in:

• cmux NEXT nightly 1.0.0-nightly.4106606310901 (08e8a12b4d)
• cmux NEXT nightly 1.0.0-nightly.4106809918401 (7deba738d3)
• cmux NEXT nightly 1.0.0-nightly.4107172498601 (6ce1bb31ea)
• cmux NEXT nightly 1.0.0-nightly.4107277115701 (57fa372ffc)
• cmux NEXT nightly 1.0.0-nightly.4107322527101 (f287294861)