Chat with an agent

Run an agent turn: evaluate rules, retrieve across attached knowledge, run the tool-calling loop, and return the answer with citations and a trace.

Authorization

TokenAuth

Authorization<token>

Project API key as Token lb-sk_...

In: header

Path Parameters

name*string

Request Body

application/json

TypeScript Definitions

Use the request body type in TypeScript.

messages*array<>

top_k?integer

Default8

conversation_id?string

Thread this turn into a conversation (groups runs; filter /runs by it).

Lengthlength <= 128

memory?

Memory hook: recall this user's memories before answering and persist the exchange after (scoped to this agent + project).

Response Body

`application/json`

curl -X POST "https://example.com/v1/agents/string/chat" \  -H "Content-Type: application/json" \  -d '{    "messages": [      {        "role": "user",        "content": "What is the status of order A123?"      }    ]  }'

{  "message": {    "role": "user",    "content": "string"  },  "citations": [    {      "text": "string",      "score": 0,      "document_id": "string",      "filename": "string"    }  ],  "ui_actions": [    {}  ],  "escalated": true,  "usage": {    "tokens_in": 0,    "tokens_out": 0  },  "trace": {    "matched_rules": [      {        "id": "string",        "rule_type": "escalation",        "condition": "string",        "action": "string",        "priority": 0      }    ],    "retrieved": [      {        "text": "string",        "score": 0,        "document_id": "string",        "filename": "string"      }    ],    "available_tools": [      "string"    ],    "tool_calls": [      {        "tool": "string",        "arguments": "string",        "result": "string",        "status": "ok"      }    ],    "plan": [      "string"    ],    "flow_path": [      "string"    ],    "input_verdict": {},    "output_verdict": {}  }}

Empty