Claude Code v2.1.271
Released Sep 14, 2026 (1 month ago).
54 issues reference this build
46 still open
8 resolved (15%)
"References this build" means the issue body mentions running 2.1.271 (from the
bug-report template). It reflects report volume, not a proven defect count — an issue may mention a
version without being caused by it. First-party data from the anthropics/claude-code tracker.
Top issues reported on v2.1.271
Ranked by community engagement (reactions and comments).
- Desktop built-in browser can't grant standing permissions for .local hosts (WordPress Studio sites)
- Desktop app (macOS): restore the copy button on the file pane
- [BUG] Claude is no longer able to sign-in into a website not even locally
- [BUG] Desktop app stealth auto-update relaunches while sessions are hosted for Remote Control, dropping every remote connection
- [BUG] Claude Desktop 2.110.0 (Windows/MSIX): after a quit, every launch fails with renderer "launch-failed, exitCode: 21" until CoworkVMService is restarted
- [Bug] Background task cleanup with sudo kills entire process tree instead of target process group
- [FEATURE] Long Monitor watches > 30-minutes without a model turn to re-arm
- [BUG] No disk-sourced skills load in 2.1.271 — user skills and enabled-plugin skills both absent; only bundled skills register
- "Approaching weekly usage limit" banner fires at 54% weekly usage
- [BUG] Reaching the 5-hour session limit kills in-flight workflow agents instead of pausing them
- [BUG] Desktop: plugin MCP OAuth sign-in page silently not opened (denied_no_activation) if the confirm dialog isn't answered within ~5s
- [Bug] Cyber security filter incorrectly blocks proposal generation
- [BUG] Linux Desktop: OAuth login callback dropped as "repeat delivery", making /login recovery impossible
- Windows: Claude Desktop Terminal panel leaks one interactive ConPTY pwsh.exe (~85 MB) per Code session; ending the session does not terminate it, and closing the tab leaks its conhost.exe
- Headless service account cannot run `claude -p`: silent exit 0 when the OS user has no onboarding state, despite `auth status` reporting `loggedIn: true`
- [BUG] Session history silently deleted with no warning when app is reopened after 30+ days
- [Bug] iOS Simulator input actions fail with sandbox: "Failed to initialize IO ports for device IO client"
- [BUG] MSIX build: claude.exe / Cowork VM bundle / VM-internal claude binary all fail to land after download — ENOENT on final rename
- [BUG] Desktop app chat sidebar stuck empty after forced re-authentication (right after chat/collaboration merge announcement) — data confirmed intact on claude.ai
- OAuth token exchange fails with HTTP 400 on fresh /login and setup-token (VS Code Remote Tunnel to Linux)
- [BUG] Adobe for Creativity plugin: OAuth sign-in always fails (gateway 403s break discovery; DCR tokens rejected)
- [BUG] MCP OAuth: DCR client_name "Claude Code (<server>)" is rejected by servers that validate RFC 7591 metadata strictly (Calendly MCP → 400 invalid_client_metadata); recurrence of #59445 / #73293, still present in 2.1.271
- [Bug] Claude Code not responding to input/commands
- Workflow docs: "default to omitting opts.model" makes every bare agent() run on the session model — the most expensive tier
- [Bug] Safety Filter False Positive on Legitimate Code Task
- [Bug] Space key conflict in agent view reply box prevents text input for new agent line
- Desktop app bundles Intel-only node-pty prebuild, triggering macOS 27 "will not open in macOS 28" warning
- Credential-handling rule: sessions disagree on whether clicking Login on a password-manager-filled form is prohibited
- Permission prompts never fire in desktop-app session; mode silently reverts to acceptEdits each turn
- [Bug] `/compact` creates duplicate live sessions operating on same directory
Changelog
What's changed
- Added fast mode in Claude Code Remote sessions (cloud and self-hosted runners): the host's fast-mode setting or
/fasttyped in the session applies where your organization allows it - Added mouse support to the
/configpanel in fullscreen mode: the wheel scrolls the settings list, a click on a setting's value changes it, and the row under the pointer is highlighted - Added
claude self-hosted-runner --drain-marker-file <path>: when that file exists at a SIGTERM drain, the runner reports its exit to the server as a host drain (telemetry only) - Added per-command
allowed_domainsto Bash, PowerShell and Monitor in auto mode with sandboxing: the hosts a command needs are reviewed with it and opened for it alone; other hosts are refused - Added
omitClaudeMdto agent frontmatter and--agentsJSON, letting custom and plugin subagents run without user, project and local CLAUDE.md files; managed policy files still load - Added
--accept-command <sha256>toclaude plugin installandclaude plugin updateto accept exactly the command a previous--jsonrun displayed, instead of-y - Added support for a
multiplierabove 1, up to 10, in themodelPricingmanaged setting and the Claude apps gatewaypricingblock, for marked-up internal chargeback rates - Added a spinner tip pointing Bedrock, Vertex AI, Foundry and LLM gateway users to the Claude desktop app; the claude.ai desktop app tip now suggests
/desktop, which offers to download the app - Fixed a cached organization policy being reused after switching accounts, organizations, or API keys, and the policy not refreshing until the hourly check when the credential changes mid-session
- Fixed the tool and command lists not updating when the organization policy finishes loading after startup or changes mid-session
- Fixed an enterprise
managed-mcp.jsonthat can't be read or parsed being ignored: it now keeps exclusive MCP control (user, project and plugin servers don't load) and warns at startup - Fixed org policy being fetched through, and rejected by, third-party local proxies set via
ANTHROPIC_UNIX_SOCKET; they are again treated like other custom gateways, including for Remote Control - Fixed cloud sessions rejecting every subagent tool call ("updatedInput … failed schema validation") when a workflow or agent approval was applied after the session's worker restarted
- Fixed
/fast offanswering "Fast mode unavailable" instead of turning fast mode off when the organization has fast mode disabled - Fixed sessions started with
CLAUDE_CODE_SKIP_FAST_MODE_ORG_CHECKre-sending fast requests every turn after the API rejected fast mode; the rejection now stands and its reason is shown - Fixed fast mode under
CLAUDE_CODE_RETRY_WATCHDOGfailing the turn on a usage-credits limit, or retrying an overload at fast speed, instead of falling back to standard speed - Fixed Bash permission checks missing the file that
fmt,columnand similar commands read when it follows an option the checker doesn't recognize - Fixed Bash permission checks skipping files a wildcard expands to when the wildcard sits in a command's pattern or option value (for example
grep -v dir/* file) - Fixed Bash permission checks so that shell variable declaration flags cannot misrepresent the command being run
- Fixed Bash commands with two directory changes, a subshell, or a
cd+gitchain skipping the prompt underpermissions.blockReadsOutsideWorkingDirectoriesin bypass and auto mode - Fixed a stale
.git/config.lockbreakinggit checkout -b,git push -uandgit configfor the rest of a session after a sandboxed command failed to start (Linux) - Fixed settings file changes made outside the session going unnoticed on macOS machines whose system file-event service is saturated; the watcher now falls back to polling
- Fixed resumed
claude -psessions whose tools all come from MCP servers failing with "At least one tool must have defer_loading=false" - Fixed tur