Anthropic: Claude Code v2.1.186 brings MCP login/logout, workflow filtering and skills management
Claude Code v2.1.186 is an update released on June 22, 2026 that introduces MCP login/logout commands for non-interactive authentication in SSH environments, status filtering in the workflow overview, and a new Skills section in the plugin interface.
This article was generated using artificial intelligence from primary sources.
Anthropic releases Claude Code v2.1.186 with MCP authentication
Anthropic released Claude Code v2.1.186 on June 22, 2026 — an update that brings functional extensions for development teams using Claude Code in automated and server environments. Unlike the previous version v2.1.185, which was exclusively a UX patch (changing the stream-stall message and raising the threshold from 10 to 20 seconds), v2.1.186 introduces new commands and expands the plugin interface.
What is MCP login/logout and why does it matter for SSH?
MCP (Model Context Protocol) — the open standard that lets AI agents access external tools and data sources — previously required interactive authentication. The new login and logout commands solve this problem: SSH sessions, CI/CD pipelines, and automated workflows can now authenticate MCP servers without user interaction. This is a key step for teams running Claude Code as part of unattended agentic pipelines.
Workflow filtering and skills management
New status filtering (key f) in the /workflows agent detail view allows development teams to quickly navigate long workflow lists — especially useful when working with multiple parallel agents. A Skills section has been added to the /plugin Installed tab, centralizing the overview and management of all installed skills in one place rather than the previous scattered approach.
Bugfixes and integrations
Version 2.1.186 resolves several bugs that affected agentic workflows: a streaming crash after the sleep command, incorrect positioning of subagent transcripts, and a permission prompt for named subagents. A fix for enterprise cost display was also added. On the integrations side, Claude Code gains iTerm2 support with fallback warnings for users running terminals without iTerm2. Bash commands prefixed with ! now automatically trigger a Claude response — behavior that is configurable according to team preferences.
The rapid delivery pace continues
Anthropic continues its fast release cycle: v2.1.185 was published on June 20 and v2.1.186 just two days later. Cumulatively, from v2.1.128 (released May 5, 2026) to v2.1.186, 58 versions shipped in fewer than seven weeks — an average of more than one release per day, reflecting the intense pace of development in the Claude Code ecosystem.
Frequently Asked Questions
- What do MCP login/logout commands mean for SSH users?
- Claude Code v2.1.186 introduces non-interactive authentication for MCP (Model Context Protocol) servers — SSH sessions and automated CI/CD workflows can now authenticate MCP without user interaction, eliminating a blocker for unattended environments.
- How does v2.1.186 differ from the previous version v2.1.185?
- While v2.1.185 was exclusively a UX patch (stream-stall message), v2.1.186 delivers functional extensions: MCP authentication, workflow filtering, a Skills section in plugins, and iTerm2 integration — along with a series of bugfixes for agentic workflows.
Related news
arXiv:2606.19992: Beyond Static Endpoints — ToolPro introduces executable programs as an interface for agentic web services
AWS: Bedrock AgentCore Payments — AI agents pay APIs autonomously via the x402 protocol
GitHub: Claude Code as an agent provider in JetBrains IDEs — public preview and cloud agent GA