Topic

glossary

Coverage, reference pages, tools, and guides connected to this topic.

  1. Agent

    A system that decides which actions to take by combining a model with tools and memory.

  2. Handoff

    The transfer of control or state from one agent to another, or from an agent to a human.

  3. Long-horizon task

    A task spanning many steps over hours or days, requiring durable state and memory.

  4. Model Context Protocol (MCP)

    An open protocol for exposing tools and context to LLMs through a standard interface.

  5. Multi-agent system

    A system of two or more agents that exchange messages or hand off tasks.

  6. Planner–critic architecture

    A pattern where a planner proposes steps and a critic prunes or revises them.

  7. Replay-based evaluation

    Scoring agent candidates against captured real-world sessions with held-out outcomes.

  8. Retrieval-augmented generation (RAG)

    Retrieving documents at inference time and conditioning generation on them.

  9. Sandbox

    An isolated execution environment for running agent code or browser actions safely.

  10. Tool use

    The pattern of an LLM invoking external functions to gather data or take action.