Claude Code claude mcp add-json stdin support
Claude Code now accepts MCP server JSON from standard input using `claude mcp add-json <name> -`. This keeps sensitive tokens off the command line.
Loading launches from 4 labs
Friday 20 Nov · Anthropic, OpenAI, Google, Meta
One card per launch from Anthropic, OpenAI, Google and Meta: what changed, how to try it, and every source linked.
Or sign in to save launches, follow topics and join discussions.
Claude Code now accepts MCP server JSON from standard input using `claude mcp add-json <name> -`. This keeps sensitive tokens off the command line.
Anthropic added fast mode support for Claude on Vertex AI across global, US, and EU regions, with automatic fallback to standard speed if fast mode is refused.
Claude Code's remote-control command now accepts a --settings file flag that applies configuration to every session it starts. Changes to the file require a restart to take effect.
New CLI command identifies latency bottlenecks in Claude API code and tests faster configurations against your evals.
A new `permissions.denyInsteadOfPrompting` setting automatically denies tool calls that would otherwise pause for permission prompts, letting the turn continue without interruption.
Claude Code now backgrounds MCP tool calls immediately and reports results when they finish, with an option to disable this behavior.
Claude Code now lets you name a session's billing group using the attributionProject setting. This helps organizations using Claude apps gateways track usage by project.
Claude Code cloud sessions on Anthropic-hosted environments now include a No repository option. Users can start a hosted cloud session without selecting a repository first.
Anthropic now applies a text watermark to content generated by Claude 4.5 and 4.6 models on supported platforms when using default sampling parameters.
Claude cloud sessions now display dedicated link cards when an MCP server asks to visit a URL. This lets users explicitly open or decline requested links rather than missing the prompt.
Claude Code cloud sessions now display form cards when an MCP server requests input, allowing users to submit answers or decline rather than having requests canceled unseen.
Developers can now sign in to a Claude apps gateway directly from the VS Code chat panel without using a terminal.
Anthropic added guardrail.input_tagging to Claude apps gateway Bedrock upstreams. It wraps prompts in the input tags required by Amazon Bedrock guardrail prompt attacks filters.
Anthropic added managed.policy_mode: layered to the Claude apps gateway. Every policy matching a developer now applies and merges over the ones below it, instead of only the first matching policy.
Anthropic added pooled group budgets to the Claude apps gateway. Groups can now share a single monthly budget on top of each member's individual limit.
Claude Code v2.1.312 adds a Bash editor with syntax colors, line numbers, and a Copy button to the setup script box in cloud environment settings. It makes reading, editing, and copying cloud setup scripts easier.
Claude Code v2.1.312 adds an effort level option to the model picker when creating or editing cloud routines. This lets you configure effort alongside the model for routines that run in the cloud.
Claude Code v2.1.312 adds click to send for queued messages in fullscreen mode. You can now click the 'to send now' hint to send a queued message right away instead of pressing its key.
The Claude apps gateway now supports oidc.user_id_claim, letting organizations identify users with an id_token claim other than sub, such as Microsoft Entra ID's oid.