Claude Code's 11 September release: plugin evals, output-style switching, and a broad reliability sweep
Claude Code ships plugin eval scoring, /output-style switching, Bash-edit diffs, and dozens of reliability fixes in its 11 September release.
Topic
22 posts about MCP from Simon Carter.
Claude Code ships plugin eval scoring, /output-style switching, Bash-edit diffs, and dozens of reliability fixes in its 11 September release.
Claude Code v2.1.257 makes Fable 5.1 the default model, adds a live /diff panel, surfaces headless commands, and fixes a batch of bugs.
Claude Code now starts before the sandbox loads, the Linux binary shrinks from ~340 MB to ~75 MB, and /cost, /usage, and /tasks get per-command token detail.
Anthropic's MHS research preview lets AI agents discover and safely control microscopes, robotic arms, and quantum hardware through a single open protocol.
Codex v0.147.0 ships portable Agent Plugins with cross-catalog search, the latest MCP spec, Amazon Bedrock caching, and smarter approval workflows.
Gemini Spark is now in beta on macOS for US Google AI Ultra subscribers, bringing desktop file automation, remote task control, and new third-party integrations.
OpenAI's GA Secure MCP Tunnel uses an outbound-only tunnel-client daemon so private MCP servers can serve ChatGPT, Codex, and the Responses API with no inbound firewall rules.
Claude Code Artifacts gain live MCP connector data, public sharing links, editor roles, and Claude Tag integration — here's what changed and what it means.
The fifth MCP spec drops stateful sessions for a request/response core, hardens OAuth/OIDC auth, and graduates MCP Apps and Tasks as versioned extensions.
Claude's new beta feature lets Team and Enterprise admins push MCP connectors to users via Okta, eliminating per-user OAuth flows on first login.
Microsoft's Power Apps MCP Server brings AI agents into model-driven apps with human-in-the-loop supervision, automated data entry, and a redesigned agent feed.
Anthropic ships three Claude Managed Agents platform updates giving enterprise developers more infrastructure control, runtime flexibility, and prompt-cache transparency.
Anthropic splits agent orchestration from tool execution, letting enterprises run agent 'hands' in their own sandboxes while Claude's 'brain' stays on Anthropic.
Codex CLI v0.139.0 lets code mode call web search directly, including from nested JS tool calls, and fixes oneOf/allOf schema preservation for MCP tools.
Claude's new public-beta dashboard gives MCP connector owners usage metrics, error diagnostics, and latency data — plus a direct submission flow to the connectors directory.
Microsoft's Work IQ API reached GA on June 16, bringing an agent-ready intelligence layer for Microsoft 365 with consumption-based Copilot Credits billing.
The latest Codex CLI update adds session archiving, OSC 8 hyperlinks in TUI markdown, richer MCP status, and short-lived WebSocket tokens for remote control.
Claude Code's latest release adds enterprise version enforcement, a /plugin list command, hook improvements, and fixes a pile of Windows and terminal rendering issues.
Anthropic's new three-tier Services Track and Partner Hub give enterprises a credentialed way to find qualified Claude implementation partners.
Claude Code renames its dynamic-workflow trigger to 'ultracode', fixes a Bash parallel batch bug, tidies the MCP panel, and adds OTEL resource-attribute labels.
Claude Code gains filesystem-first plugin loading from .claude/skills, a scaffolding command, autocomplete, and hot-reload — no marketplace needed.
Anthropic has acquired Stainless, the startup behind every official Claude SDK and MCP server generator, in a deal reportedly worth over $300M.