[ca] CLI Updates: Claude Code 2.1.139, Copilot 1.0.45, Codex 0.130.0, GitHub MCP v1.0.4, Playwright Browser v1.60.0 #31644

Claude Code is published on NPM only and does not have a public GitHub repository, so detailed per-version release notes are not available. The version progression spans 6 patch releases over 4 days indicating active bug-fix iteration.

Package Links

• NPM Package: https://www.npmjs.com/package/`@anthropic-ai/claude-code`

Impact Assessment

• Risk: Low
• Patch-level releases with no documented breaking changes

Bug Fixes (1.0.45)

• Sessions with extension permission prompts can now be resumed without "Session file is corrupted" error
• agentStop hook now fires correctly when the agent stops via task_complete

Bug Fixes (1.0.44)

• Path completion in /add-dir no longer flickers or gets intercepted by @ and # pickers
• Shell commands via ! prefix work correctly with all shell configurations; shell aliases and rc file settings now work in ! commands
• Quota display correctly shows remaining usage for Free users instead of always showing 100% used
• Tool permissions granted in autopilot mode are preserved after /clear
• Effort level applies correctly when switching models via the /model picker
• Pressing Ctrl+C while a permission prompt is pending no longer causes the CLI to hang
• Project info remains visible in slash command picker when no results match
• Invalid URL entries in settings.json no longer crash CLI startup; they are skipped with a warning
• Timeline shows the resolved model for rubber-duck sub-agents (e.g. Rubber-duck(claude-opus-4.7))

CLI Discovery

• New top-level slash commands: /autopilot, /fork
• New flag: --prerelease for copilot update and /update

Bug Fixes

• Live app-server threads pick up config changes without requiring a restart (app-server: refresh live threads from latest config snapshot openai/codex#21187)
• Turn diffs stay accurate across apply-patch operations, including partial failures that still mutated files (Make turn diff tracking operation backed openai/codex#21180, fix: preserve exact turn diffs after partial apply_patch failures openai/codex#21518)
• Thread summaries, renames, resume, and fork paths work better through ThreadStore, including threads without local rollout paths (Move thread name edits to ThreadStore openai/codex#21264, Route ThreadManager rollout path reads through thread store openai/codex#21265, [codex] Fix pathless thread summaries openai/codex#21266)
• Remote compaction emits response.processed for v2 streams and avoids sending service_tier on API-key compact requests (Send response.processed after remote compaction v2 openai/codex#21642, Omit service_tier from remote /responses/compact requests under API auth openai/codex#21676)
• Windows sandbox setup grants sandbox users access to the desktop runtime binary cache (Grant sandbox users access to desktop runtime bin openai/codex#21564)
• Removed stale "research preview" wording from the codex exec startup banner (Remove exec research preview banner wording openai/codex#21683)

Chores

• Configurable OpenTelemetry trace metadata; richer review/feedback analytics for better debugging and triage
• Hardened dependency and CI hygiene with fully qualified GitHub Action pins, a Dependabot cooldown, and a cargo-shear upgrade
• Simplified internal surfaces by removing unused device-key APIs, extra skills roots, remote thread-store implementation, and string-keyed MCP tool maps

Notable PRs

• add top-level remote-control command openai/codex#21424 — add top-level remote-control command
• Show plugin hooks in plugin details openai/codex#21447 — Show plugin hooks in plugin details
• feat: make built-in MCPs first-class runtime servers openai/codex#21356 — make built-in MCPs first-class runtime servers
• feat(app-server, threadstore): Thread pagination APIs and ThreadStore contract openai/codex#21566 — Thread pagination APIs and ThreadStore contract
• feat: enable AWS login credentials for Bedrock auth openai/codex#21623 — enable AWS login credentials for Bedrock auth

New APIs

Browser, Context and Page

• Event browser.on('context') — fired when a new context is created
• BrowserContext now mirrors lifecycle events from its pages: download, frameattached, framedetached, framenavigated, pageclose, pageload

Locators and Assertions

• New description option in getByRole() for matching the accessible description
• New pseudo option in expect(locator).toHaveCSS() reads computed styles from ::before or ::after
• New style option in locator.highlight() applies extra inline CSS; new page.hideHighlight() to clear all highlights

Network

• webSocketRoute.protocols() returns the WebSocket subprotocols requested by the page
• New noDefaults option in browserType.connectOverCDP() disables Playwright's default overrides (download behavior, focus emulation, media emulation)

Errors and Reporting

• New webError.location() mirrors consoleMessage.location()
• consoleMessage.location() now exposes line / column properties (lineNumber / columnNumber are deprecated)
• New testInfoError.errorContext surfaces additional diagnostic context
• reporter.onError() now receives a workerInfo argument for fixture teardown errors

Test runner

• New {testFileBaseName} token in testProject.snapshotPathTemplate
• Test runner now errors when a config tries to override a non-option fixture, and rejects workers: 0 or negative values

Other Improvements

• HTML reporter: npx playwright show-report accepts .zip files directly; steps with nested attachments show an indicator
• Trace Viewer adds a pretty-print toggle for JSON / form request and response bodies in the network details panel

Browser Versions

• Chromium 148.0.7778.96
• Mozilla Firefox 150.0.2
• WebKit 26.4

1. Constants updated in pkg/constants/version_constants.go:

   • DefaultClaudeCodeVersion: 2.1.133 → 2.1.139
   • DefaultCopilotVersion: 1.0.43 → 1.0.45
   • DefaultCodexVersion: 0.129.0 → 0.130.0
   • DefaultGitHubMCPServerVersion: v1.0.3 → v1.0.4
   • DefaultPlaywrightBrowserVersion: v1.59.1 → v1.60.0

2. Workflow lock files regenerated via make recompile (219 workflows updated)

3. Testing recommendations:

   • Run make build && make test to verify code compiles cleanly
   • Run smoke tests: smoke-claude, smoke-copilot, smoke-codex workflows
   • For Playwright workflows: verify no usage of removed deprecated APIs (ariaRef, exposeBinding handle option, connect logger option, videosPath, videoSize)
   • Verify Copilot CLI: MCPs load correctly and /models works on PATs (per existing constant pin comment)