Blog/Tags/tool-calling

tool-calling

Browse 8 articles tagged with “tool-calling”.

Articles tagged “tool-calling”

8 articles

Warm watercolor illustration of an MCP client and server exchanging a request, a paused input form, and a re-submitted call in a loop

MCP now uses round-trips, not streams, for user input

The July 28 spec deprecated elicitation and sampling. Multi Round-Trip Requests (SEP-2322) replace both with a stateless pattern: the server returns InputRequiredResult, the client collects input, the client re-submits. Here's how to migrate.

Warm watercolor illustration of an AI agent following DNS records and a well-known signpost to locate an MCP server

Production MCP·13 min read

How AI agents discover your MCP server automatically

A practical guide to publishing .well-known/mcp.json and DNS TXT records so any AI agent can find and connect to your MCP server without manual configuration.

Warm watercolor illustration of a person watching flowing streams of light on a live display

Protocol Deep Dives·15 min read

MCP tasks: how async tool calls fix your agent's timeout problem

The November 2025 MCP spec introduced Tasks: a call-now, fetch-later primitive that lets agents kick off slow operations without blocking. Here's how it works and how to build with it.

Warm watercolor illustration of four colored bands connecting an AI agent to tools, sub-agents, enterprise systems, and external partner agents

Protocol Deep Dives·15 min read

Four protocols that power every AI agent

MCP handles tools, A2A handles agent delegation, ACP routes across enterprise frameworks, and ANP crosses org boundaries. Here's how the four protocols divide the work and when each belongs in your stack.

Warm watercolor illustration of the MCP 2026 specification unfolding into four branches for extensions, tasks, apps, and authorization

Protocol Deep Dives·12 min read

MCP's 2026 spec changes what your tools can do

The MCP spec RC is out with Extensions, Tasks, MCP Apps, and OAuth hardening. Here's what changes for AI agent builders before the July 28 final.

Warm watercolor illustration of a developer wiring an MCP client and server together, with three labeled channels for sampling, roots, and elicitation

Protocol Deep Dives·12 min read

How to Use MCP Sampling, Roots, and Elicitation in AI Agents

Most MCP tutorials cover only server-side features: tools, resources, prompts. The three client capabilities (Sampling, Roots, Elicitation) unlock human-in-the-loop patterns that server tools alone can't. Here's how to use them.

Warm watercolor illustration of a librarian robot squinting at a wall of drawer labels, several smudged or peeling, while holding a request slip

Testing & Evaluation·13 min read

How MCP Tool Descriptions Break Your Agent

New research shows 97% of MCP tool descriptions have quality issues that hurt agent accuracy. Here's what the smells look like, why they matter, and how to fix them.

Warm watercolor illustration of a chat window blooming open like a pop-up book to reveal a small interactive calendar card with a hand reaching in to tap a slot

Tools & MCP·14 min read

MCP Apps: Build UIs That Render Inside AI Chat

MCP Apps let your tools return interactive HTML dashboards, forms, and visualizations that render inline in Claude, ChatGPT, and VS Code. Here's how to build them for your agent tools.

MCP changelog and dev notes

Short, occasional updates on the MCP spec, new Protobox features, and patterns we've seen in production. No marketing fluff.

Be among the first