Claude Code v2.1.274：修复MCP连接超时、会话状态丢失及多项Bug

What's changed

• Added a visible warning when memory usage is critical, with steps to free memory or restart safely
• Added CLAUDE_CODE_MCP_STARTUP_WAIT_MS to bound how long the first non-interactive turn waits for connecting MCP servers (0 = don't wait)
• Added effort attribute to the claude_code.llm_request OpenTelemetry trace span, matching the api_request event
• Added claude_code.managed_settings_resolved OTel event: managed-settings sources and policy helper state; redacted settings and digests with OTEL_LOG_MANAGED_SETTINGS=1
• Added store.connect_timeout_seconds to the Claude apps gateway config to lengthen the Postgres connect timeout (default 5 seconds), and improved the boot error when the database is unreachable to point to store.postgres_url and the configured timeout
• Added enduser.sub, the IdP subject, to the telemetry Claude Desktop and Cowork send through a Claude apps gateway
• Added a Claude apps gateway warning when a replica has more requests open than the 256 it sends upstream at once, and a startup log line showing that limit
• Added click-to-expand for collapsed teammate and agent messages in fullscreen mode
• Fixed sessions getting stuck endlessly retrying "unexpected tool_use_id" 400 errors: corrupted transcripts now self-heal where possible, and otherwise a clear error (with a /rewind hint) ends the loop
• Fixed MCP servers configured as http that only speak legacy HTTP+SSE failing to connect when they answer the first request with 422 or another 4xx error
• Fixed Streamable HTTP MCP tool calls timing out after about 5 minutes even when a longer per-server timeout was set
• Fixed MCP prompts and resources not refreshing when a server sends list-changed notifications without declaring listChanged
• Fixed MCP tool calls refused with 403 insufficient_scope being reported as an expired sign-in: the error now names the missing permissions and points to /mcp re-authentication
• Fixed hook-driven sessions (such as an active /goal) ending with "Prompt is too long" instead of compacting when the context overflowed again after a reactive compaction
• Fixed an active /goal being lost when resuming (--continue / --resume) a session that had compacted
• Fixed claude agents losing --model, --effort, --permission-mode, --allow-dangerously-skip-permissions and --agent after an auto-update relaunch
• Fixed a per-turn slowdown when a language server publishes project-wide diagnostics for thousands of files
• Fixed subagents with model: "opus" on Bedrock, Vertex or Foundry leaving the session's model when its id has no recognizable model family (unless ANTHROPIC_DEFAULT_OPUS_MODEL is set)
• Fixed self-hosted runner sessions failing every turn with a 401 after a few failed token refreshes, until the next scheduled refresh; the runner now keeps retrying, and fetches a new token after a 401
• Fixed clickable links to local file paths doing nothing in VS Code and other terminals that require a file:// URI
• Fixed the transcript renumbering ordered lists in your own messages (typing "3. 2. 1." displayed "3. 4. 5."); numbers and "N)" markers now show as typed
• Fixed AskUserQuestion preview notes being attached to a previously chosen option instead of the highlighted one
• Fixed AskUserQuestion preview mode dropping the highlighted option when submitting a note with Enter
• Fixed a resumed background agent keeping half of an interrupted tool batch when one of its calls was approved with a message
• Fixed a local claude -p --resume started with CLAUDE_CODE_RESUME_INTERRUPTED_TURN not reporting background tasks the previous process left unfinished
• Fixed the first turn of a cloud session sometimes starting without the tools of an SDK-hosted MCP server that was still connecting
• Fixed background agent notifications claiming the agent had no live background work when it was still waiting on its own background task and would resume
• Fixed error hints in Claude Desktop sessions to suggest slash commands like /usage-credits instead of CLI flags that cannot be used there
• Fixed /schedule saving a routine's prompt without its message role when Claude writes the routine in the shape that listing routines returns
• Fixed /status not showing the apiKeyHelper failure that its own error banner told you to check
• Fixed /fast on in non-interactive sessions reporting on and then turning off under an organization's managed fast mode policy; it now says the organization has disabled it
• Fixed the Artifact tool asking you to approve an update to an artifact that it then refused because the session had not read the latest version
• Fixed Cowork and claude.ai cloud sessions with network access on treating reads of a teammate's artifact as if network access were off
• Fixed a plugin or marketplace directory with no git repository of its own taking its version from an enclosing git repository, such as a git-managed ~/.claude
• Fixed --strict-mcp-config with an empty --mcp-config holding the first non-interactive turn for up to MCP_TIMEOUT on incidental MCP servers
• Fixed Stop prompt hooks re-sending their whole prompt on every block in a conversation; repeat blocks now name the condition with a 500-character label
• Fixed extra empty editor windows opening at startup on Linux under Wayland when running inside the Cursor or VS Code terminal
• Fixed an unhandled promise rejection in the Claude apps gateway when Postgres drops a connection during a spend check
• Fixed Claude apps gateway cutting every open stream on SIGTERM: it now lets in-flight requests finish for up to 25 seconds before exiting (CLAUDE_GATEWAY_DRAIN_TIMEOUT_MS)
• Fixed installed_plugins.json being rewritten on nearly every start-up when plugin policy comes from remote managed settings, which made Claude Desktop reload every open session's plugins
• Fixed headless and SDK sessions making a separate model call for every background task that finished; completions already queued are now answered by one call
• Fixed the Bash tool re-sourcing the shell profile (a multi-second stall on the next command) after every plugin reload; it now does so only when the plugins' bin/ directories changed
• Fixed plugins with a top-level $schema in hooks/hooks.json showing an "unknown key" notice
• Fixed MCP connection errors and the MCP login tool's description showing secrets resolved from ${VAR} placeholders in MCP configs
• Fixed Bash permission checks for commands that loop over or assign certain special shell variables; these commands now ask for permission
• Fixed worktree-isolated sessions accepting Bash commands with certain nested shell expansions; these are now refused
• Fixed the Edit permission prompt preview sometimes showing a different location than the approved edit in files with multi-byte characters
• Fixed background commands being stopped after 30 idle minutes on machines under mild memory pressure; they're now stopped only when memory is critically low, and the debug log says why
• Fixed a message a subagent sends to the main session disappearing from the Claude Desktop transcript after a relaunch
• Fixed a plugin loaded from a .zip being served from a stale extraction after several overlapping reloads
• Fixed a sub-agent's progress summary being replaced by a runaway multi-paragraph reply
• Improved startup in --input-format stream-json sessions: the first turn no longer waits up to 2s for still-connecting MCP servers whose tools tool search defers; they arrive on a later turn
• Improved Monitor tool notifications: a script's final output and its exit now arrive as one notification instead of two, saving a model turn
• Improved Artifact tool errors: when you are not signed in to claude.ai the terminal now says so on the first attempt, and Claude is told to stop retrying a rejected call sooner
• Improved artifact publishing: a publish built on an older version is stopped before it is sent, with the newer page to merge