[maria-rcks]

Problem

Bash line editors can treat carriage-return-separated setup wrappers as unfinished multiline input. Worktree setup then never reports completion and the waiting agent cannot start.

Change

Submit complete POSIX commands with encoded script text, preserve trailing newlines, and evaluate in a subshell before reporting exit status. Long scripts use bounded, independently complete assignments. Preserve Fish, PowerShell, and main's startup-readiness handling. Reuse the existing test harness instead of adding another mock runner.

Scope and approval

This is an obvious, focused bug fix under the small-fix contribution exception. It only repairs POSIX completion-observed setup and settle delivery, preserving the original intent and all other shell paths. Maintainer takeover was already recorded on https://git.cafe/pingdotgg/t3code/pulls/17817, and Maria explicitly requested this replacement after the archived fork rejected pushes.

Verification

• vp run check:changed passed through Blacksmith: format, lint, both knip passes, GraphQL generation, affected typechecks, and all 12 adjacent tests. The six shell cases include literals/heredocs, long input, stdin, trailing backslash-newline, explicit exit, and syntax error. Two fail against current main before the fix.
• Real dev client with Bash 5.2 and ble.sh: main remains in MULTILINE with setup waiting; this branch completes the same setup and the Codex agent replies setup verified.
• Additional real node-pty checks pass with plain Bash and ble.sh: multiline/heredoc, trailing backslash-newline, explicit exit 7, syntax error 2, and 120-line Unicode script.
• Two independent GPT-6.1-Sol static reviews found no blockers. The replacement's file contents are identical to the reviewed merge candidate.
• Unverified: native zsh, Fish, PowerShell, macOS/Windows, mobile/Electron, relay/tunnel. Long-input byte limits are tested; the long case was not repeated through the browser. Existing cosmetic limit: line-editor redraws can show internal assignments for long scripts.

Takeover

Replaces https://git.cafe/pingdotgg/t3code/pulls/17817, with thanks to @sgaabdu4 for the original implementation.

The contributor fork is archived and rejects maintainer pushes. Compared with the original PR's current diff, this retains its production wrapper and all six shell regression cases, replaces the duplicate test harness with main's harness, and preserves main's startup-readiness tests. No behavior was dropped. The commit credits Abid A Gafoor.

Built with GPT-6.1-Sol in the Codex harness on T3 Code.

[macroscopeapp]

Approvability

Verdict: Approved at d1c9df2

Macroscope's review found this PR approvable — POSIX setup and settle scripts now use complete, size-bounded shell commands so line editors can execute them and report completion. The change preserves existing script selection and execution authority, with no product-default changes.

^{You can add or adjust custom eligibility rules. Learn more.}

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	5.0 KiB	5.0 KiB	0 B (0.0%)	6.8 KiB	✅
Codex	Thread snapshot wire	3.8 KiB	3.8 KiB	0 B (0.0%)	4.9 KiB	✅
Codex	Live turn WebSocket wire	1.2 KiB	1.2 KiB	0 B (0.0%)	2.0 KiB	✅
Codex	Live turn WebSocket decoded	21.5 KiB	21.5 KiB	0 B (0.0%)	24.4 KiB	✅
Codex	Live turn messages	1	1	0 (0.0%)	8	✅
Claude	Total thread wire	5.0 KiB	5.0 KiB	0 B (0.0%)	6.8 KiB	✅
Claude	Thread snapshot wire	3.8 KiB	3.8 KiB	0 B (0.0%)	4.9 KiB	✅
Claude	Live turn WebSocket wire	1.2 KiB	1.2 KiB	0 B (0.0%)	2.0 KiB	✅
Claude	Live turn WebSocket decoded	21.9 KiB	21.9 KiB	0 B (0.0%)	24.4 KiB	✅
Claude	Live turn messages	1	1	0 (0.0%)	8	✅

Other scenarios

All 64 scenario measurements match main.

All 64 scenario measurements

Provider	Scenario	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	short turn resume	WebSocket decoded	10.5 KiB	10.5 KiB	0 B (0.0%)	12.7 KiB	✅
Codex	short turn resume	WebSocket messages	2	2	0 (0.0%)	2	✅
Codex	long turn running, cold open	HTTP decoded	281.7 KiB	281.7 KiB	0 B (0.0%)	302.7 KiB	✅
Codex	long turn running, cold open	rows	75	75	0 (0.0%)	75	✅
Codex	long turn running, history page	HTTP decoded	92.4 KiB	92.4 KiB	0 B (0.0%)	107.4 KiB	✅
Codex	long turn running, history page	rows	75	75	0 (0.0%)	75	✅
Codex	long turn finished, cold open	HTTP decoded	381.5 KiB	381.5 KiB	0 B (0.0%)	410.2 KiB	✅
Codex	long turn finished, cold open	rows	151	151	0 (0.0%)	151	✅
Codex	long turn finished, socket cold open	WebSocket decoded	381.6 KiB	381.6 KiB	0 B (0.0%)	410.2 KiB	✅
Codex	long turn finished, socket cold open	WebSocket messages	2	2	0 (0.0%)	2	✅
Codex	long turn finished, socket cold open	rows	151	151	0 (0.0%)	151	✅
Codex	long turn finished, work page	HTTP decoded	186.2 KiB	186.2 KiB	0 B (0.0%)	200.2 KiB	✅
Codex	long turn finished, work page	rows	150	150	0 (0.0%)	150	✅
Codex	idle thread resume	WebSocket decoded	65 B	65 B	0 B (0.0%)	200 B	✅
Codex	idle thread resume	WebSocket messages	1	1	0 (0.0%)	1	✅
Codex	fleet shell snapshot	HTTP decoded	42.4 KiB	42.4 KiB	0 B (0.0%)	53.7 KiB	✅
Codex	fleet shell snapshot	rows	27	27	0 (0.0%)	27	✅
Codex	fleet shell subscribe	WebSocket decoded	65 B	65 B	0 B (0.0%)	200 B	✅
Codex	fleet shell subscribe	WebSocket messages	1	1	0 (0.0%)	1	✅
Codex	fleet shell resume check	HTTP decoded	18 B	18 B	0 B (0.0%)	100 B	✅
Codex	fleet shell resume check	rows	0	0	0	0	✅
Codex	stale shell reload	HTTP decoded	42.4 KiB	42.4 KiB	0 B (0.0%)	53.7 KiB	✅
Codex	stale shell reload	rows	27	27	0 (0.0%)	27	✅
Codex	stale shell resume	WebSocket decoded	140 B	140 B	0 B (0.0%)	400 B	✅
Codex	stale shell resume	WebSocket messages	2	2	0 (0.0%)	2	✅
Codex	stale todo list cold open	HTTP decoded	72.4 KiB	72.4 KiB	0 B (0.0%)	83.0 KiB	✅
Codex	stale todo list cold open	rows	75	75	0 (0.0%)	75	✅
Codex	stale todo list cold open	nodes	2	2	0 (0.0%)	4	✅
Codex	stale todo list cold open	subagents	2	2	0 (0.0%)	4	✅
Codex	stale todo list cold open	plans	1	1	0 (0.0%)	2	✅
Codex	stale todo list cold open	providerTurns	1	1	0 (0.0%)	2	✅
Codex	stale todo list cold open	checkpoints	1	1	0 (0.0%)	2	✅
Claude	short turn resume	WebSocket decoded	10.6 KiB	10.6 KiB	0 B (0.0%)	12.7 KiB	✅
Claude	short turn resume	WebSocket messages	2	2	0 (0.0%)	2	✅
Claude	long turn running, cold open	HTTP decoded	283.3 KiB	283.3 KiB	0 B (0.0%)	302.7 KiB	✅
Claude	long turn running, cold open	rows	75	75	0 (0.0%)	75	✅
Claude	long turn running, history page	HTTP decoded	93.4 KiB	93.4 KiB	0 B (0.0%)	107.4 KiB	✅
Claude	long turn running, history page	rows	75	75	0 (0.0%)	75	✅
Claude	long turn finished, cold open	HTTP decoded	384.2 KiB	384.2 KiB	0 B (0.0%)	410.2 KiB	✅
Claude	long turn finished, cold open	rows	151	151	0 (0.0%)	151	✅
Claude	long turn finished, socket cold open	WebSocket decoded	384.3 KiB	384.3 KiB	0 B (0.0%)	410.2 KiB	✅
Claude	long turn finished, socket cold open	WebSocket messages	2	2	0 (0.0%)	2	✅
Claude	long turn finished, socket cold open	rows	151	151	0 (0.0%)	151	✅
Claude	long turn finished, work page	HTTP decoded	188.1 KiB	188.1 KiB	0 B (0.0%)	200.2 KiB	✅
Claude	long turn finished, work page	rows	150	150	0 (0.0%)	150	✅
Claude	idle thread resume	WebSocket decoded	66 B	66 B	0 B (0.0%)	200 B	✅
Claude	idle thread resume	WebSocket messages	1	1	0 (0.0%)	1	✅
Claude	fleet shell snapshot	HTTP decoded	43.0 KiB	43.0 KiB	0 B (0.0%)	53.7 KiB	✅
Claude	fleet shell snapshot	rows	27	27	0 (0.0%)	27	✅
Claude	fleet shell subscribe	WebSocket decoded	66 B	66 B	0 B (0.0%)	200 B	✅
Claude	fleet shell subscribe	WebSocket messages	1	1	0 (0.0%)	1	✅
Claude	fleet shell resume check	HTTP decoded	18 B	18 B	0 B (0.0%)	100 B	✅
Claude	fleet shell resume check	rows	0	0	0	0	✅
Claude	stale shell reload	HTTP decoded	43.0 KiB	43.0 KiB	0 B (0.0%)	53.7 KiB	✅
Claude	stale shell reload	rows	27	27	0 (0.0%)	27	✅
Claude	stale shell resume	WebSocket decoded	140 B	140 B	0 B (0.0%)	400 B	✅
Claude	stale shell resume	WebSocket messages	2	2	0 (0.0%)	2	✅
Claude	stale todo list cold open	HTTP decoded	73.4 KiB	73.4 KiB	0 B (0.0%)	83.0 KiB	✅
Claude	stale todo list cold open	rows	75	75	0 (0.0%)	75	✅
Claude	stale todo list cold open	nodes	2	2	0 (0.0%)	4	✅
Claude	stale todo list cold open	subagents	2	2	0 (0.0%)	4	✅
Claude	stale todo list cold open	plans	1	1	0 (0.0%)	2	✅
Claude	stale todo list cold open	providerTurns	1	1	0 (0.0%)	2	✅
Claude	stale todo list cold open	checkpoints	1	1	0 (0.0%)	2	✅

Baseline: 2f2730c · PR result: d1c9df2 · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 108.5 KiB
• Claude decoded thread snapshot: 108.8 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.

[coderabbitai]

No actionable comments were generated in the recent review.

Walkthrough

The POSIX completion wrapper now encodes setup commands, splits long commands into terminal-sized lines, evaluates them in a subshell, and appends a status sentinel. Tests check command submission, multiline syntax, literal text, long scripts, trailing backslash-newline behavior, and exit codes.

Changes: POSIX setup command delivery

Layer / File(s)	Summary
POSIX wrapper and execution validation apps/server/src/project/ProjectSetupScriptRunner.ts, apps/server/src/project/ProjectSetupScriptRunner.test.ts	The wrapper encodes commands for printf %b and splits encoded wrappers that exceed 1,023 UTF-8 bytes. It evaluates the command in a subshell and appends a sentinel with the exit status. Tests run submitted commands through Bash and check multiline syntax, literal text, long scripts, trailing backslash-newline behavior, and exit codes.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Change: Bug fix

Suggested reviewers: t3dotgg, juliusmarminge

Merge Risk: Minimal · up to d1c9d

This change makes setup scripts submit as complete commands so Bash line editors no longer leave them waiting, and it keeps exit-status reporting intact. No concrete defect was found, so it looks ready to merge.

Pre-merge checks: 4

Check name	Status	Explanation
Title check		The title names the main change: setup scripts now complete when a shell line editor (Bash with ble.sh) is active. The raw summary shows the POSIX completion wrapper was rewritten to submit complete c…
Description check		The description follows the repository template in substance. It has Problem, Change, Scope, and Verification sections. The Problem section explains the bug and its effect. The Change section describe…
Linked Issues check		Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check		Check skipped because no linked issues were found for this pull request.

Full details: Title check

Explanation

The title names the main change: setup scripts now complete when a shell line editor (Bash with ble.sh) is active. The raw summary shows the POSIX completion wrapper was rewritten to submit complete commands and append a sentinel, which matches the title. The title is specific and follows conventional commit style.

Full details: Description check

Explanation

The description follows the repository template in substance. It has Problem, Change, Scope, and Verification sections. The Problem section explains the bug and its effect. The Change section describes the fix. The Scope section explains why no issue is linked and cites the maintainer request. The Verification section lists focused shell tests, real-client checks, and what was not verified. The description does not link a bug issue in the usual format. It links an earlier PR and a fork, which is not the same as an issue. It also includes a takeover section that the template does not require. These gaps are minor, so the check passes.

Finishing Touches

Generate unit tests (beta)

• Commit to this branch
• Create a new PR

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Recent review info

Run configuration

• Source mode: Auto
• Repository: pingdotgg/t3code
• Branch: maria/takeover-pr-17817
• File: .coderabbit.config.ts
• Commit: d1c9df2
• Layers: Repository TS config
• Review profile: CHILL
• Plan: Advanced
• Run ID: b56db014-68d9-429c-9162-76d6c71cb5ee

Commits

Reviewing files that changed from the base of the PR and between 8c07e7f and d1c9df2.

Files selected for processing: 2

• apps/server/src/project/ProjectSetupScriptRunner.test.ts
• apps/server/src/project/ProjectSetupScriptRunner.ts

Included review availability: This review used your included allowance. Your plan provides up to 10 included reviews per hour; 8 remain after this review.

[maria-rcks]

Note

Written by gpt-6.1-sol on behalf of Maria

Verified in a real dev environment

verified unchanged head 301c1a1a30b00a7ef3bae145ef3c6c310a0f8ff7 on this linux boat vm. git.cafe and github both use https://git.cafe/pingdotgg/t3code/pulls/25472. the pr is open; its sole commit has no authors from the excluded maintainer accounts.

• installed dependencies with vp i in a separate github/main worktree (4914cdd98f) and on the exact pr head. ran both dev environments in the background with separate state. used repository playwright and /usr/bin/google-chrome.
• created a project pointing at a real git repository through the client. configured a synchronous worktree setup action through the action editor. used bash 5.2.21 with ble.sh cb9c2c3.
• on main, the multiline script with an apostrophe, quoted heredoc, unicode, and a trailing comment stays in ble.sh's MULTILINE state. setup waits and the provider does not start.
• on this head, the same script completes, creates setup-proof.txt, and starts a real codex gpt-6.1-sol turn through the local proxy. codex runs cat setup-proof.txt and returns setup verified.
• neighboring behavior: a 120-line unicode/quoted/backslash script completes through the same worktree flow, then codex reads its proof file. exit 7 reports exactly that status in the client and prevents the agent from starting.
• blacksmith vp run check:changed 071239804e375078dd3fae3b14704b4def57e048 passed format, lint, both knip passes, graphql generation, affected typechecks, and all 12 adjacent tests. an explicit pr base was needed because the worker has shallow history. the pr currently has 29 successful checks, none failing or pending.

short runtime transcript:

$ git rev-parse HEAD
301c1a1a30b00a7ef3bae145ef3c6c310a0f8ff7
$ cat .t3/worktrees/setup-verification/t3-795aa242/setup-proof.txt
setup verified
$ cat .t3/worktrees/setup-verification/t3-d20de44d/setup-proof.txt
setup verified
client, exit-7 scenario: Setup script exited with 7.
blacksmith adjacent tests: 12 passed

Not verified

native ios/android and electron clients, macos/windows, zsh/fish/powershell, relay/tunnel connections, other providers, and the settle-trigger path were not exercised on this linux web-client pass. interactive stdin and trailing backslash-newline have passing adjacent automated coverage, but were not repeated through the browser. no performance claim is made; ble.sh redraws expose internal assignments during long scripts.

[AkiVonAkira]

Note

🤖 Claude Opus 5.5 responding on behalf of Kayzel

Reproduced on T3 Code 0.1.7-nightly.20261102.4061 (server on Linux, Bash 5.3.9, ble.sh 0.4.0-nightly+fb1811e). A worktree setup script stayed in ble.sh's -- MULTILINE -- (RET or C-m: insert a newline, C-j: run) state, the step sat on "Seed Unity Library" for several minutes, and the agent never started. The script itself runs in about 10 ms when invoked directly.

Typing the current POSIX wrapper (( cmd\r); printf '\n__T3_SETUP_DONE__%s\n' "$?"\r) into an interactive Bash with ble.sh attached reproduces it. With ble.sh left detached, the same input prints the sentinel. I'm working around it by skipping ble-attach when NO_COLOR=1 and FORCE_COLOR=0, which only setup terminals set. I'd like to drop that once this lands.

Dismissing prior approval to re-evaluate d1c9df2

[juliusmarminge]

Evidence

Created a fresh demo repository and configured the reported multiline Setup action in the UI with automatic worktree execution and waiting enabled. Started the same worktree thread through the CLI on both commits. Environment: Linux, Chromium 1280×800, light theme, Bash 5.2 with ble.sh attached, fresh databases, replay agent.

Proven: base remains in MULTILINE with Setup running and Start agent pending after 60 seconds; head completes setup and releases the agent, which finishes its reply. scenarios:prove exited 0.

	Before	After
Commit	2f2730c113	d1c9df20f9
Result
Recording	Before recording	After recording

Scenario source · Shell fixture driver · Runner report

Not shown:

• Nonzero-exit reporting, long chunked commands, stdin-reading scripts, and other shells were not exercised; this proof covers the reported synchronous ble.sh success hang.