Repository navigation
fix(bg): match background-session command args on token boundaries - #1834
Conversation
commandLineContainsArgs matched each stored launch arg as a raw substring, so a selector like "1642" was satisfied by an unrelated live token "16420" in the same position. A reused PID whose command line merely contained the stored digits could therefore keep a dead session classified as running, and the kill path could target the wrong process. Match each stored arg against a whole whitespace-delimited token, in order, instead of a substring. Add regression coverage through isBackgroundSessionProcessAlive for the collision, the exact-token match, the session-id match, and the dead-process case. Refs Twigpine#1770
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: ASSERTIVE Plan: Pro Run ID: 📒 Files selected for processing (2)
📜 Recent review details⏰ Context from checks skipped due to timeout. (2)
🧰 Additional context used📓 Path-based instructions (4)src/**/*.{ts,tsx}📄 CodeRabbit inference engine (AGENTS.md)
Files:
**⚙️ CodeRabbit configuration file
Files:
**/*⚙️ CodeRabbit configuration file
Files:
{src/**/*.test.ts,src/**/*.test.tsx,tests/**,scripts/**/*.test.ts,vscode-extension/**/*.test.js}⚙️ CodeRabbit configuration file
Files:
🔇 Additional comments (3)
📝 WalkthroughWalkthroughBackground session liveness checks now use whole-token command matching instead of substring scans, and tests cover exact matches, collision cases, multi-word argv handling, and dead-process rejection. ChangesBackground session identity matching fix
Estimated code review effort: 2 (Simple) | ~10 minutes Suggested labels: Suggested reviewers: 🚥 Pre-merge checks | ✅ 5 | ❌ 2❌ Failed checks (2 warnings)
✅ Passed checks (5 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 3
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/cli/bgRegistry.test.ts`:
- Around line 681-732: Add a companion regression test in
isBackgroundSessionProcessAlive process identity that covers the sessionId
matching path, since it is still vulnerable to substring collisions. Use the
existing BackgroundSession fixture and verify a command line containing a longer
unrelated token like sess-100 does not match sessionId sess-1, while keeping the
exact-token case positive. Refer to isBackgroundSessionProcessAlive and the
sessionId/commandLine matching branch in bgRegistry.ts so the new test guards
the hardened tokenized behavior.
In `@src/cli/bgRegistry.ts`:
- Around line 507-529: The `commandLineMatchesBackgroundSession` path still uses
a raw `includes` check for `session.sessionId`, so it can misclassify a process
when the session ID appears as a substring inside an unrelated token. Update
that sessionId lookup to use the same whole-token matching approach as
`commandLineContainsArgs` (or share the tokenization logic) so `sessionId`
matching is consistent with the fixed `session.command` path and avoids the
`1642` vs `16420` collision.
- Around line 514-526: The argument matching in bgRegistry.ts’s session liveness
check is splitting commandLine on whitespace, which breaks multi-word argv
values like refactor auth. Update the matching logic in the arg/commandLine
comparison block to preserve quoted or free-text argv entries as single
arguments, and make sure the loop in the relevant matcher still compares each
arg against the correct token sequence without treating spaces inside a value as
separators.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
Run ID: d944096c-2aa3-4b18-86c8-d7ee280cc3b6
📒 Files selected for processing (2)
src/cli/bgRegistry.test.tssrc/cli/bgRegistry.ts
📜 Review details
⏰ Context from checks skipped due to timeout. (2)
- GitHub Check: smoke-and-tests
- GitHub Check: typecheck
🧰 Additional context used
📓 Path-based instructions (9)
src/**/*.{ts,tsx}
📄 CodeRabbit inference engine (AGENTS.md)
Use TypeScript with strict mode and ESM imports
Files:
src/cli/bgRegistry.tssrc/cli/bgRegistry.test.ts
**/*.{ts,tsx,js,jsx,py,json,md}
📄 CodeRabbit inference engine (CONTRIBUTING.md)
Follow the existing code style in the touched files
Files:
src/cli/bgRegistry.tssrc/cli/bgRegistry.test.ts
**/*.{ts,tsx,js,jsx,py}
📄 CodeRabbit inference engine (CONTRIBUTING.md)
Keep comments useful and concise
Files:
src/cli/bgRegistry.tssrc/cli/bgRegistry.test.ts
**/*.{ts,tsx}
📄 CodeRabbit inference engine (CONTRIBUTING.md)
Follow TypeScript strict mode and type safety practices by running typecheck before submitting
Files:
src/cli/bgRegistry.tssrc/cli/bgRegistry.test.ts
**
⚙️ CodeRabbit configuration file
**: # AGENTS.md - AI Agent Coding GuideThis guide is for AI coding agents working in the OpenClaude repository. Read it before changing code, and also follow CONTRIBUTING.md for contributor policy, PR expectations, review follow-up, and project scope.
Project Snapshot
OpenClaude is a coding-agent CLI for cloud and local model providers. It supports OpenAI-compatible APIs, Anthropic, Gemini, DeepSeek, Ollama, MCP, local backends, slash commands, tools, agents, and a React/Ink terminal UI.
The installed CLI runs on Node.js
>=22.0.0. Bun is used for source builds, scripts, dependency management, and tests.Work Style
- Keep changes focused on one problem.
- Prefer existing patterns in the file or nearby module.
- Avoid unrelated formatting, renames, dependency changes, or broad rewrites.
- Add or update tests when behavior changes.
- Update docs when setup, commands, provider behavior, or user-facing behavior changes.
- For new features, larger refactors, dependencies, or runtime changes, follow the issue-first guidance in CONTRIBUTING.md.
Stack And Conventions
- TypeScript with strict mode and ESM imports.
- React + Ink for terminal UI.
- Bun lockfile and Bun scripts for development workflows.
- Node runtime for the built CLI.
Common libraries and patterns:
chalkfor terminal color.commanderfor CLI argument parsing.execafor child processes.- Existing service, provider, settings, permission, and UI patterns over new abstractions.
Repository Map
src/commands/- slash and CLI command implementations.src/components/- React/Ink UI components.src/services/- API, MCP, OAuth, wiki, voice, and other service integrations.src/tools/- tool implementations.src/utils/- shared utilities.src/integrations/- provider and model integration metadata.src/entrypoints/- CLI, MCP, SDK, and generated public types.src/tasks/- local, remote, workflow, and monitor tas...
Files:
src/cli/bgRegistry.tssrc/cli/bgRegistry.test.ts
**/*
⚙️ CodeRabbit configuration file
**/*: Apply the OpenClaude maintainer review rubric from AGENTS.md. Review the current diff, not stale discussion context. Separate real blockers from suggestions. Do not request changes for vague style churn. Treat approval as merge-ready from CodeRabbit's side, pending required human review and GitHub Checks. If checks are failing or unavailable, say so clearly instead of implying the PR is fully ready.
Files:
src/cli/bgRegistry.tssrc/cli/bgRegistry.test.ts
**/*.test.{ts,tsx,js,jsx}
📄 CodeRabbit inference engine (CONTRIBUTING.md)
Add or update tests when the change affects behavior
Files:
src/cli/bgRegistry.test.ts
**/*.test.{ts,tsx,js}
📄 CodeRabbit inference engine (CONTRIBUTING.md)
Test the exact provider/model path you changed when possible
Files:
src/cli/bgRegistry.test.ts
{src/**/*.test.ts,src/**/*.test.tsx,tests/**,scripts/**/*.test.ts,vscode-extension/**/*.test.js}
⚙️ CodeRabbit configuration file
{src/**/*.test.ts,src/**/*.test.tsx,tests/**,scripts/**/*.test.ts,vscode-extension/**/*.test.js}: Review tests for meaningful coverage of the changed behavior, isolation of global/env/config state, async cleanup, fake timers, provider profile leaks, and Windows-compatible assumptions. Block when risky runtime changes lack focused regression coverage or tests assert implementation details while missing the user-visible behavior.
Files:
src/cli/bgRegistry.test.ts
🔇 Additional comments (2)
src/cli/bgRegistry.ts (1)
531-539: LGTM! (unchanged; see the sessionId note attached to lines 507-529 for the follow-up)src/cli/bgRegistry.test.ts (1)
9-15: LGTM!
jatmn
left a comment
jatmn
left a comment
There was a problem hiding this comment.
Summary
CodeRabbit's requested changes have been open for more than five hours, and this repository's CONTRIBUTING.md says PR authors should address CodeRabbit findings instead of waiting for a maintainer override. I verified the current patch, and the session-id and multi-word argv comments are still valid, so please complete those review items before this moves forward.
Findings
-
[P1] Complete CodeRabbit's request to harden the session-id match
src/cli/bgRegistry.ts:535
The new matcher fixes substring collisions forsession.command, but the first branch still returnstruewhen the live command merely containssession.sessionIdas a substring. For example, a stored id likesess-1would still match an unrelated live command containingsess-100, leaving that reused PID classified as the background session before thekillpath decides whether to signal it. Please complete CodeRabbit's request by routing the session-id check through the same exact/whole-token identity matching instead ofincludes. -
[P2] Complete CodeRabbit's request to preserve multi-word argv values
src/cli/bgRegistry.ts:514
Background prompts are stored as argv entries, and--from-pr/resume launches can rely onsession.commandbecause the resolved session id is not present on argv. A real prompt such asrefactor authis stored as one arg, butpsrenders the live command as words like... --from-pr 1642 --print refactor auth; after this changesplit(/\s+/)producesrefactorandauth, so it can no longer match the stored argrefactor auth. The scanner also keeps skipping later tokens until it finds a match, so it still does not compare the stored argv sequence to the live command sequence reliably. That can mark a live background PR-resume session stale, or keep trusting a command where the matching token appears later in the wrong argument position. Please update the matcher so it preserves argv boundaries and compares each stored arg against the correct live argument position.
… args Address review on Twigpine#1770: - commandLineMatchesBackgroundSession matched session.sessionId with a raw includes(), so a short id like 'sess-1' matched an unrelated live command containing 'sess-100' and kept a reused-PID session classified as running. Route it through the same whole-token matcher. - commandLineContainsArgs required each stored arg to equal a single token, so a prompt stored as one argv entry (e.g. 'refactor auth') never matched the words ps renders it as, breaking --from-pr/resume launches that rely on the stored command. Expand each arg into its own tokens and match the flattened sequence as an ordered subsequence. Add regression tests: session-id substring collision stays dead, exact id token is alive, and a multi-word prompt arg matches across command tokens (all proven fail-on-old).
|
Thanks — both addressed in 34fa3a5.
Added regression tests: session-id substring collision stays dead, the exact id token is alive, and a multi-word prompt arg matches across command tokens. Verified all three fail on the pre-fix matcher and pass after. |
There was a problem hiding this comment.
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
src/cli/bgRegistry.ts (1)
507-538: 🔒 Security & Privacy | 🟠 Major | ⚡ Quick winNon-contiguous subsequence matching reopens a collision gap this PR is meant to close.
commandLineContainsArgsmatchesargTokensas an ordered subsequence, not a contiguous block — the innerwhileloop skips forward over any non-matching live tokens between matches. That means a live command likenode x openclaude y 1642would satisfy stored args['node', 'openclaude', '1642'], because each token is still found in order, just with unrelated tokens interspersed. A reused PID whose command line happens to contain the same ordered tokens with extra junk between them will still be misclassified as the owning session — the same class of false-positive collision#1770targets, just via token insertion instead of digit-substring overlap.None of the added tests cover this interspersed-token case, so it isn't caught by the new regression suite either.
🔧 Suggested fix: require contiguous match
- if (argTokens.length === 0) return false - let tokenIndex = 0 - for (const argToken of argTokens) { - let matched = false - while (tokenIndex < tokens.length) { - const token = tokens[tokenIndex] - tokenIndex += 1 - if (token === argToken) { - matched = true - break - } - } - if (!matched) return false - } - return true + if (argTokens.length === 0 || argTokens.length > tokens.length) return false + outer: for (let start = 0; start <= tokens.length - argTokens.length; start++) { + for (let i = 0; i < argTokens.length; i++) { + if (tokens[start + i] !== argTokens[i]) continue outer + } + return true + } + return false🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/cli/bgRegistry.ts` around lines 507 - 538, The `commandLineContainsArgs` helper in `bgRegistry` still allows non-contiguous ordered subsequence matches, which can misclassify unrelated live processes as matching a stored session. Change the matching logic so `args` must appear as one contiguous run of whole tokens in `commandLine` rather than skipping over intervening tokens, and keep the whitespace-tokenization behavior for multi-word args. Update or add a regression test around `commandLineContainsArgs` to cover an interspersed-token case that should not match.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Outside diff comments:
In `@src/cli/bgRegistry.ts`:
- Around line 507-538: The `commandLineContainsArgs` helper in `bgRegistry`
still allows non-contiguous ordered subsequence matches, which can misclassify
unrelated live processes as matching a stored session. Change the matching logic
so `args` must appear as one contiguous run of whole tokens in `commandLine`
rather than skipping over intervening tokens, and keep the
whitespace-tokenization behavior for multi-word args. Update or add a regression
test around `commandLineContainsArgs` to cover an interspersed-token case that
should not match.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
Run ID: 54185d9e-e17b-4a7b-80ad-c17bf45cfa87
📒 Files selected for processing (2)
src/cli/bgRegistry.test.tssrc/cli/bgRegistry.ts
📜 Review details
⏰ Context from checks skipped due to timeout. (2)
- GitHub Check: typecheck
- GitHub Check: smoke-and-tests
🧰 Additional context used
📓 Path-based instructions (9)
src/**/*.{ts,tsx}
📄 CodeRabbit inference engine (AGENTS.md)
Use TypeScript with strict mode and ESM imports
Files:
src/cli/bgRegistry.tssrc/cli/bgRegistry.test.ts
**/*.{ts,tsx,js,jsx,py,json,md}
📄 CodeRabbit inference engine (CONTRIBUTING.md)
Follow the existing code style in the touched files
Files:
src/cli/bgRegistry.tssrc/cli/bgRegistry.test.ts
**/*.{ts,tsx,js,jsx,py}
📄 CodeRabbit inference engine (CONTRIBUTING.md)
Keep comments useful and concise
Files:
src/cli/bgRegistry.tssrc/cli/bgRegistry.test.ts
**/*.{ts,tsx}
📄 CodeRabbit inference engine (CONTRIBUTING.md)
Follow TypeScript strict mode and type safety practices by running typecheck before submitting
Files:
src/cli/bgRegistry.tssrc/cli/bgRegistry.test.ts
**
⚙️ CodeRabbit configuration file
**: # AGENTS.md - AI Agent Coding GuideThis guide is for AI coding agents working in the OpenClaude repository. Read it before changing code, and also follow CONTRIBUTING.md for contributor policy, PR expectations, review follow-up, and project scope.
Project Snapshot
OpenClaude is a coding-agent CLI for cloud and local model providers. It supports OpenAI-compatible APIs, Anthropic, Gemini, DeepSeek, Ollama, MCP, local backends, slash commands, tools, agents, and a React/Ink terminal UI.
The installed CLI runs on Node.js
>=22.0.0. Bun is used for source builds, scripts, dependency management, and tests.Work Style
- Keep changes focused on one problem.
- Prefer existing patterns in the file or nearby module.
- Avoid unrelated formatting, renames, dependency changes, or broad rewrites.
- Add or update tests when behavior changes.
- Update docs when setup, commands, provider behavior, or user-facing behavior changes.
- For new features, larger refactors, dependencies, or runtime changes, follow the issue-first guidance in CONTRIBUTING.md.
Stack And Conventions
- TypeScript with strict mode and ESM imports.
- React + Ink for terminal UI.
- Bun lockfile and Bun scripts for development workflows.
- Node runtime for the built CLI.
Common libraries and patterns:
chalkfor terminal color.commanderfor CLI argument parsing.execafor child processes.- Existing service, provider, settings, permission, and UI patterns over new abstractions.
Repository Map
src/commands/- slash and CLI command implementations.src/components/- React/Ink UI components.src/services/- API, MCP, OAuth, wiki, voice, and other service integrations.src/tools/- tool implementations.src/utils/- shared utilities.src/integrations/- provider and model integration metadata.src/entrypoints/- CLI, MCP, SDK, and generated public types.src/tasks/- local, remote, workflow, and monitor tas...
Files:
src/cli/bgRegistry.tssrc/cli/bgRegistry.test.ts
**/*
⚙️ CodeRabbit configuration file
**/*: Apply the OpenClaude maintainer review rubric from AGENTS.md. Review the current diff, not stale discussion context. Separate real blockers from suggestions. Do not request changes for vague style churn. Treat approval as merge-ready from CodeRabbit's side, pending required human review and GitHub Checks. If checks are failing or unavailable, say so clearly instead of implying the PR is fully ready.
Files:
src/cli/bgRegistry.tssrc/cli/bgRegistry.test.ts
**/*.test.{ts,tsx,js,jsx}
📄 CodeRabbit inference engine (CONTRIBUTING.md)
Add or update tests when the change affects behavior
Files:
src/cli/bgRegistry.test.ts
**/*.test.{ts,tsx,js}
📄 CodeRabbit inference engine (CONTRIBUTING.md)
Test the exact provider/model path you changed when possible
Files:
src/cli/bgRegistry.test.ts
{src/**/*.test.ts,src/**/*.test.tsx,tests/**,scripts/**/*.test.ts,vscode-extension/**/*.test.js}
⚙️ CodeRabbit configuration file
{src/**/*.test.ts,src/**/*.test.tsx,tests/**,scripts/**/*.test.ts,vscode-extension/**/*.test.js}: Review tests for meaningful coverage of the changed behavior, isolation of global/env/config state, async cleanup, fake timers, provider profile leaks, and Windows-compatible assumptions. Block when risky runtime changes lack focused regression coverage or tests assert implementation details while missing the user-visible behavior.
Files:
src/cli/bgRegistry.test.ts
🔇 Additional comments (4)
src/cli/bgRegistry.ts (2)
540-547: LGTM! sessionId now correctly routed through whole-token matching, closing the gap flagged in the prior review round.
553-577: LGTM!src/cli/bgRegistry.test.ts (2)
6-16: LGTM!
725-771: LGTM! Good regression coverage for the sessionId whole-token and multi-word prompt cases requested in prior review. Note: the interspersed-token/non-contiguous matching gap flagged inbgRegistry.ts(lines 507-538) isn't covered by this suite either — worth adding once that's addressed.
jatmn
left a comment
jatmn
left a comment
There was a problem hiding this comment.
Summary
CodeRabbit's latest requested change is still valid, and it has been open for more than five hours. This repository's CONTRIBUTING.md says PR authors should address CodeRabbit findings instead of waiting for a maintainer override, so please complete that automated review item before this moves forward.
Findings
- [P1] Complete CodeRabbit's request to require a contiguous command match
src/cli/bgRegistry.ts:527
The new matcher still treats the stored background command as an ordered subsequence of the live command line, so it skips over unrelated tokens between matches. For example, a stored command like['node', 'openclaude', '1642']is still classified as alive for a reused PID whose command line isnode attacker openclaude extra 1642 --serve, because the loop advances pastattackerandextraand eventually finds each stored token. That leaves the same wrong-processkillrisk open for token insertion collisions, and the added tests only cover1642vs16420, not this interspersed-token case. Please complete CodeRabbit's request by requiring the flattened stored argv tokens to appear as one contiguous whole-token run in the live command line, and add a regression test where unrelated tokens inserted between the stored tokens are rejected.
…en run The token-boundary matcher for background-session process identity treated the stored argv tokens as an ordered subsequence of the live command line, so unrelated tokens between matches were skipped. A reused PID whose command interleaves the stored tokens (e.g. stored [node, openclaude, 1642] against "node attacker openclaude extra 1642 --serve") was still classified alive, keeping the wrong-process kill risk open for token-insertion collisions. Match the flattened stored argv as one contiguous whole-token run instead. Leading interpreter path and trailing flags still match; interspersed tokens no longer do. Add a regression test asserting an interspersed-token command line is rejected (fails on the prior subsequence matcher, passes now). Refs Twigpine#1770
|
Addressed in 529de15. The matcher now requires the flattened stored argv tokens to appear as one contiguous whole-token run, not an ordered subsequence — so the interspersed case you called out ( Added a regression test ( |
jatmn
left a comment
jatmn
left a comment
There was a problem hiding this comment.
I found an issue that needs to be addressed before this is ready.
Findings
- [P2] Preserve Windows background sessions that rely on command fallback
src/cli/bgRegistry.ts:526
The new matcher splits both the live command line and the stored argv on whitespace, but WindowsGet-CimInstance ... CommandLinereturns the raw command line with quoted paths and quoted arguments, whilesession.commandstores unquoted argv entries such asC:\Program Files\nodejs\node.exeorrefactor auth. For non-forked--from-prresumes,buildBackgroundSessionLaunch()deliberately does not add the resolved session id to argv, so this command fallback is the only identity path. A real Windows command like"C:\Program Files\nodejs\node.exe" C:\repo\dist\cli.mjs --from-pr 1642 --print "refactor auth"no longer matches stored['C:\Program Files\nodejs\node.exe', 'C:\repo\dist\cli.mjs', '--from-pr', '1642', '--print', 'refactor auth'], sorefreshBackgroundSessionStatuses()marks a live session stale andopenclaude kill <id>can report it killed without signalling the process. Please make the token-boundary fix handle quoted Windows command lines, or avoid requiring the quoted executable/entrypoint/prompt tokens before falling back to the argv selector tokens.
Get-CimInstance CommandLine returns the raw, quoted Windows command line, so
a whitespace split fuses quotes onto the edge tokens ("C:\Program,
node.exe", "refactor, auth") while session.command stores those values
unquoted. The contiguous whole-token run therefore never matched, and a live
non-forked --from-pr resume (whose only identity path is the stored command)
was marked stale, letting kill <id> report success without signalling the
process. Trim a single surrounding quote from each token before comparing;
POSIX ps output is unquoted so this is a no-op there and preserves the Twigpine#1770
token-boundary guard.
|
Good catch on the Windows path — took the quote-handling route.
Fix trims a single surrounding quote from each token before comparing ( Added two tests: the exact quoted Windows line from your example (spaced exe path + quoted |
jatmn
left a comment
jatmn
left a comment
There was a problem hiding this comment.
Thanks for the update. I rechecked the previously discussed background-session identity paths and do not see any remaining actionable issues from my side.
@kevincodex1 LGTM
4 participants
Summary
commandLineContainsArgs(src/cli/bgRegistry.ts) matched each stored launch arg as a raw substring viaindexOf, so a selector like1642was satisfied by an unrelated live token16420in the same position. A reused PID whose command line merely contains the stored digits could keep a dead background session classified asrunning, and thekillpath could then signal the wrong process.This is a partial/regression fix, not a full resolution — hence
Refs #1770rather thanFixes #1770. The broader process-identity work (pre-signal revalidation, durable argv marker, legacy-session fallback) is deliberately out of scope here and left for its own PRs.Impact
openclaude kill <id-or-name>and background-session liveness are less likely to act on an unrelated process that merely reuses a PID with a superset-token command line. Legitimate sessions are still recognized by exact command tokens or by the session id on the command line.commandLineContainsArgsnow tokenizes on whitespace and requires whole-token matches; behavior is unchanged for real invocations whose argv tokens match exactly.Testing
bun run buildbun run smokebun run checkbun test src/cli/bgRegistry.test.ts(27 pass)Added four cases through the exported
isBackgroundSessionProcessAlive(injectingisProcessAlive/getProcessCommand): the1642vs16420collision (nowfalse), exact-token match (true), session-id match (true), and dead process (false). The collision case is proven to fail against the previous substring implementation and pass after the change.Notes
kill, and a persisted opaque argv marker for new sessions) belong in follow-up PRs. Happy to take a run at the pre-signal revalidation next once this lands, following maintainer sequencing.Summary by CodeRabbit