{
 "id": "https://redfinchadvisory.com/lexicon/agent",
 "slug": "agent",
 "prefLabel": "Agent",
 "altLabel": [
  "AI agent"
 ],
 "hiddenLabel": [],
 "definition": "An AI system that works toward a goal by deciding its own next steps and using tools to take actions.",
 "scopeNote": "What makes something an agent is that the model, not fixed code, decides what to do next. Describe any agent with four questions: what you hand it (autonomy), whose permissions it uses (identity), what it can change (permissions), and what starts it (trigger).",
 "tier": "core",
 "group": "Agents",
 "broader": [],
 "narrower": [
  "https://redfinchadvisory.com/lexicon/coding-agent",
  "https://redfinchadvisory.com/lexicon/personal-agent",
  "https://redfinchadvisory.com/lexicon/background-agent"
 ],
 "related": [
  {
   "id": "https://redfinchadvisory.com/lexicon/assistant",
   "type": "confused"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/workflow-automation",
   "type": "confused"
  }
 ],
 "links": [
  {
   "id": "https://redfinchadvisory.com/lexicon/agent-harness",
   "label": "turns a model into",
   "direction": "in"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/a2a",
   "label": "connects agent to",
   "direction": "in"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/eval",
   "label": "test",
   "direction": "in"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/guardrail",
   "label": "filter inputs and outputs of",
   "direction": "in"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/human-on-the-loop",
   "label": "watches and can stop",
   "direction": "in"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/human-oversight",
   "label": "how people check",
   "direction": "in"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/levels-of-autonomy",
   "label": "sits on the upper rungs of",
   "direction": "out"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/agentic",
   "label": "is carried out by an",
   "direction": "in"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/agentic-workflow",
   "label": "has fixed steps unlike an",
   "direction": "in"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/multi-agent-system",
   "label": "splits one job across many",
   "direction": "in"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/tool",
   "label": "is called by an",
   "direction": "in"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/human-in-the-loop",
   "label": "approves each action of",
   "direction": "in"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/ai-washing",
   "label": "often misuses the word",
   "direction": "in"
  }
 ],
 "modified": "2026-10-21",
 "markdown": "https://redfinchadvisory.com/lexicon/agent.md",
 "inScheme": {
  "id": "https://redfinchadvisory.com/lexicon",
  "version": "1.0",
  "currentAsOf": "2026-10-21"
 }
}