50 articles in category "Agents"
May 8, 2026 • Agents, AGNTCY
The Agent Directory Service directories can be exposed to LLMs directly using the Model Context Protocol and deploying the entire stack into a production environment.
May 8, 2026 • Agents, OpenClaw
OpenClaw is an open-source, self-hosted AI agent framework that turns your machine into a persistent, autonomous personal assistant.
May 6, 2026 • Agents, AGNTCY
The Agent Directory Service is the phonebook for our agent ecosystem. However, it is not just mapping a name to an address like DNS does for web servers. Instead, ADS relies on capability-based matching.
March 20, 2026 • Microsoft Agent Framework, Agents, Production
Some tool calls are too consequential to let an agent run unsupervised. Function approvals pause the agent run when a destructive or sensitive tool is about to fire, surface the call to a human, and resume only after explicit approval. This article looks at how to mark a tool as approval-required, how the run pauses, and how to wire approve and reject back into the agent.
March 15, 2026 • Microsoft Agent Framework, Agents, Memory
AgentSession keeps one conversation alive for as long as the Python process runs. Context providers let an agent remember things across sessions, across users, and across long timescales. This article looks at the provider abstraction, the built-in history and memory providers, and how to write your own.
March 14, 2026 • Agents, AGNTCY
In a multi-agent environment, relying on unvalidated metadata is a recipe for disaster.
March 11, 2026 • Agents, AGNTCY
Without a standardized schema, orchestrators are left guessing what an agent can do, often resorting to fragile prompt parsing and heuristic-based routing. This is neither scalable nor reliable.
March 10, 2026 • Agents, AGNTCY
The AI agent ecosystem is crowded with acronyms — MCP, A2A, ACP, LangGraph, CrewAI, AutoGen. It is genuinely confusing. I hear this from practitioners constantly. Everyone is building an 'agent standard,' so how do we make sense of it all?
March 9, 2026 • Agents, AGNTCY
Building a global network of agents is not a trivial undertaking. It requires a robust, standardized foundation. Much like the OSI model formalized how computers communicate across networks, AGNTCY introduces a structured stack to handle everything from discovery to secure invocation.
March 6, 2026 • Microsoft Agent Framework, Agents
Free-text replies are fragile to parse, especially when downstream code depends on the answer's shape. This article looks at how Microsoft Agent Framework constrains an agent's reply to a Pydantic model, when to use it, and how it interacts with tool calls and streaming.
