Tag
9 posts tagged.
The desktop app and the CLI run the same engine and read the same files. On three points they resolve them in a different order, and Anthropic's own docs call one of them a departure from the CLI scope hierarchy. Read 10 September 2026.
ACP and MCP sit on opposite sides of the agent: MCP connects an agent to tools, ACP connects an editor to the agent. The spec calls itself MCP-friendly and passes your MCP servers inside session/new. The catch is the transport floor: stdio is mandatory, HTTP and SSE are optional capabilities.
The MCP 2026-07-28 revision marks the Sampling feature deprecated under SEP-2577, eight months after the revision that expanded it. A field log on what the deprecation registry actually says, why elicitation survived and sampling did not, and what to check before rewriting anything.
Elicitation has been in MCP since June 2025, but the current 2026-07-28 revision changed how it is delivered and calls the old pattern a breaking change. A field log on the three-revision diff, why elicitationId became requestState, and why the statefulness advice inverted.
MCP Apps went stable as the first official MCP extension in January 2026. A pre-build field log on the identifier mismatch, the bilateral opt-in that fails silently, the deny-by-default CSP, and which hosts actually render one in August 2026.
Every comparison ranking for this query calls goose Block's open-source agent. Block donated it to the Agentic AI Foundation at the Linux Foundation in April 2026, alongside Anthropic's MCP. A field log on what that changes and what it does not.
A working reference: the five MCP servers I keep loaded in Claude Code after a year of daily use, the ones I removed, why every server costs context-window tokens, and the exact scope config that keeps the list short.
A 30-day field log of running an MCP server in production: four failure modes the tutorials skip, honest latency and cost numbers, and when an MCP server is the wrong tool.
I wire an MCP server to my IDE's agent in about thirty minutes, and suddenly it reads my real Postgres schema and project files instead of hallucinating. MCP is just a standard way for agents to call external tools. I pick one server that solves a real annoyance, drop a small JSON config with command, args, and env, restart, and let the agent fetch its own context. That's the whole win.
