morluto/rea

Fork 0

mirror of https://github.com/morluto/rea.git synced 2026-11-05 06:45:22 +00:00

[PR #2036] [CLOSED] fix(process): name CS_RESTRICT environment withholding in ownership diagnostics #4977

New issue

Closed

opened 2026-10-31 00:57:37 +00:00 by bsakaguchi · 0 comments

bsakaguchi commented

2026-10-31 00:57:37 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/morluto/rea/pull/2036
Author: @Ghost-Kang
Created: 10/12/2026
Status: ❌ Closed

Base: main ← Head: fix/process-restricted-env-diagnostic

📝 Commits (1)

39df0e7 fix(process): name CS_RESTRICT environment withholding in ownership diagnostics

📊 Changes

3 files changed (+120 additions, -96 deletions)

View changed files

📝 bridge/process/ProcessRunTokenReader.swift (+19 -2)
📝 src/process/ProcessOwnership.test.ts (+99 -94)
📝 src/process/ProcessOwnership.ts (+2 -0)

📄 Description

Summary

Refs #2035. On macOS, process captures intermittently fail with cleanup_incomplete / environment_unavailable when an unrelated CS_RESTRICT process (Chrome or Electron helpers, etc.) starts during the capture. This PR is diagnostic only: the run-token reader now names that case restricted_process_environment_withheld instead of the generic environment_unavailable. Containment is unchanged — the candidate still fails closed. Whether to exempt such processes (as #1057 did for platform binaries) is left to the maintainer decision in #2035.

Problem and expected behavior

XNU omits the environment of cs_restricted processes from KERN_PROCARGS2. bridge/process/ProcessRunTokenReader.swift only recognised CS_PLATFORM_BINARY in its environmentUnavailable path, so restricted processes were indistinguishable from a genuinely transient read failure. Expected: the cleanup reason says why the token was unreadable.

Change and scope

Swift reader: codeSigningStatus(pid:) is factored out of isPlatformBinary and adds isRestrictedProcess (CS_RESTRICT = 0x800). In the environmentUnavailable catch, the order is platform binary → restricted → environment_unavailable. If csops fails, the result stays environment_unavailable.
ProcessOwnership.ts: sanitizedTokenReadFailure passes the new reason through, so the reason names it instead of other_unavailable.
Intentionally not changed:
- environmentWithheldFromInspection, so there is no new exemption.
- The reader's runId == nil path, where changing classification would alter ownership outcomes.
Both are discussed in #2035.

Contract and boundary impact

Semantic owner and earliest changed stage: Darwin run-token reader (cleanup diagnostics)
CLI and MCP/tool-catalog contract: the cleanup_incomplete reason text can now contain restricted_process_environment_withheld; the error code, category and shape are unchanged
Provider, bridge, target-format, or platform compatibility: macOS only; the Swift bridge gains one more csops call on an already-failing path
Evidence, artifact, provenance, or reconstruction contract: none
Process execution, authorization, cleanup, or containment impact: none; restricted candidates remain fail-closed whether related or unrelated (covered by tests)
Generated metadata (docs/public/product-catalog.json), package, or installation impact: none

Evidence and regression coverage

Tests added or updated: src/process/ProcessOwnership.test.ts. The unrelated-candidate fail-closed test and the related-candidate fail-closed test are parametrised over restricted_process_environment_withheld.
Base reproduction or other evidence: reproduction steps are in #2035. With an unrelated CS_RESTRICT process spawned every 0.3 s, the "accepts NUL" acceptance test fails on main with environment_unavailable=N. With this PR, after npm run compile, it still fails (4/4, containment unchanged), but the reason reads restricted_process_environment_withheld=N.

User-visible CLI/MCP output:

owned_process_group: process ownership token could not be read for 1 live process(es): restricted_process_environment_withheld=1; live candidates <pid>=restricted_process_environment_withheld

Remaining proof gaps: there is no automated native test that launches a real CS_RESTRICT image, because no stable restricted binary is available on CI. The Swift path was verified manually as above.

Validation performed

npx vitest run src/process tests/boundary/process/darwinProcessRunTokenReader.test.ts: 24 files, 229 passed, 1 skipped
npm run check: passed
xcrun swiftc -typecheck bridge/process/ProcessRunTokenReader.swift: passed
The manual restricted-noise repro described above
pre-commit and pre-push hooks: passed
Environment: macOS 26.6.2 arm64, Node 24.21.0, Xcode 27.0

Compatibility, safety, and release

Breaking changes or migration steps: none
Real Hopper/Ghidra, browser, or OS coverage: macOS 26.6.2 only
Package or release metadata impact: none
Security, privacy, process, or containment review: no relaxation; the new reason is fail-closed everywhere

Review checklist

This PR addresses a concrete problem or an agreed enhancement.
The PR has one focused outcome and the title follows type(scope): outcome.
Related issue is linked, or the reason for not linking one is stated above.
Tests cover changed observable behavior and meaningful failure paths.
Owning docs, contracts, and generated metadata are updated where needed.
User-visible CLI/MCP changes include representative output.
I checked the final diff for secrets, unrelated cleanup, and unsupported claims.

🤖 Generated with Claude Code

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/morluto/rea/pull/2036 **Author:** [@Ghost-Kang](https://github.com/Ghost-Kang) **Created:** 10/12/2026 **Status:** ❌ Closed **Base:** `main` ← **Head:** `fix/process-restricted-env-diagnostic` --- ### 📝 Commits (1) - [`39df0e7`](https://github.com/morluto/rea/commit/39df0e7584ec871faf7c39bd036321ff0b77f873) fix(process): name CS_RESTRICT environment withholding in ownership diagnostics ### 📊 Changes **3 files changed** (+120 additions, -96 deletions) <details> <summary>View changed files</summary> 📝 `bridge/process/ProcessRunTokenReader.swift` (+19 -2) 📝 `src/process/ProcessOwnership.test.ts` (+99 -94) 📝 `src/process/ProcessOwnership.ts` (+2 -0) </details> ### 📄 Description ## Summary Refs #2035. On macOS, process captures intermittently fail with `cleanup_incomplete` / `environment_unavailable` when an unrelated `CS_RESTRICT` process (Chrome or Electron helpers, etc.) starts during the capture. This PR is **diagnostic only**: the run-token reader now names that case `restricted_process_environment_withheld` instead of the generic `environment_unavailable`. Containment is unchanged — the candidate still fails closed. Whether to exempt such processes (as #1057 did for platform binaries) is left to the maintainer decision in #2035. ## Problem and expected behavior XNU omits the environment of `cs_restricted` processes from `KERN_PROCARGS2`. `bridge/process/ProcessRunTokenReader.swift` only recognised `CS_PLATFORM_BINARY` in its `environmentUnavailable` path, so restricted processes were indistinguishable from a genuinely transient read failure. Expected: the cleanup reason says why the token was unreadable. ## Change and scope - **Swift reader:** `codeSigningStatus(pid:)` is factored out of `isPlatformBinary` and adds `isRestrictedProcess` (`CS_RESTRICT` = `0x800`). In the `environmentUnavailable` catch, the order is platform binary → restricted → `environment_unavailable`. If `csops` fails, the result stays `environment_unavailable`. - **`ProcessOwnership.ts`:** `sanitizedTokenReadFailure` passes the new reason through, so the reason names it instead of `other_unavailable`. - **Intentionally not changed:** - `environmentWithheldFromInspection`, so there is no new exemption. - The reader's `runId == nil` path, where changing classification would alter ownership outcomes. Both are discussed in #2035. ## Contract and boundary impact - Semantic owner and earliest changed stage: Darwin run-token reader (cleanup diagnostics) - CLI and MCP/tool-catalog contract: the `cleanup_incomplete` reason text can now contain `restricted_process_environment_withheld`; the error code, category and shape are unchanged - Provider, bridge, target-format, or platform compatibility: macOS only; the Swift bridge gains one more `csops` call on an already-failing path - Evidence, artifact, provenance, or reconstruction contract: none - Process execution, authorization, cleanup, or containment impact: none; restricted candidates remain fail-closed whether related or unrelated (covered by tests) - Generated metadata (`docs/public/product-catalog.json`), package, or installation impact: none ## Evidence and regression coverage - Tests added or updated: `src/process/ProcessOwnership.test.ts`. The unrelated-candidate fail-closed test and the related-candidate fail-closed test are parametrised over `restricted_process_environment_withheld`. - Base reproduction or other evidence: reproduction steps are in #2035. With an unrelated `CS_RESTRICT` process spawned every 0.3 s, the "accepts NUL" acceptance test fails on `main` with `environment_unavailable=N`. With this PR, after `npm run compile`, it still fails (4/4, containment unchanged), but the reason reads `restricted_process_environment_withheld=N`. - User-visible CLI/MCP output: ``` owned_process_group: process ownership token could not be read for 1 live process(es): restricted_process_environment_withheld=1; live candidates <pid>=restricted_process_environment_withheld ``` - Remaining proof gaps: there is no automated native test that launches a real `CS_RESTRICT` image, because no stable restricted binary is available on CI. The Swift path was verified manually as above. ## Validation performed - `npx vitest run src/process tests/boundary/process/darwinProcessRunTokenReader.test.ts`: 24 files, 229 passed, 1 skipped - `npm run check`: passed - `xcrun swiftc -typecheck bridge/process/ProcessRunTokenReader.swift`: passed - The manual restricted-noise repro described above - pre-commit and pre-push hooks: passed - Environment: macOS 26.6.2 arm64, Node 24.21.0, Xcode 27.0 ## Compatibility, safety, and release - Breaking changes or migration steps: none - Real Hopper/Ghidra, browser, or OS coverage: macOS 26.6.2 only - Package or release metadata impact: none - Security, privacy, process, or containment review: no relaxation; the new reason is fail-closed everywhere ## Review checklist - [x] This PR addresses a concrete problem or an agreed enhancement. - [x] The PR has one focused outcome and the title follows `type(scope): outcome`. - [x] Related issue is linked, or the reason for not linking one is stated above. - [x] Tests cover changed observable behavior and meaningful failure paths. - [x] Owning docs, contracts, and generated metadata are updated where needed. - [x] User-visible CLI/MCP changes include representative output. - [x] I checked the final diff for secrets, unrelated cleanup, and unsupported claims. 🤖 Generated with [Claude Code](https://claude.com/claude-code) --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

bsakaguchi

2026-10-31 00:57:37 +00:00

closed this issue
added the
pull-request
label

bsakaguchi referenced this issue

2026-11-01 21:42:26 +00:00

[PR #4977] [CLOSED] fix(setup): remove every repeated REA-owned JSON member #6706

No milestone

No project

No assignees

1 participant

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

morluto/rea#4977

No description provided.

Rows
Columns