Skip to main content
POST
/
agents
/
generate-system-prompt
Generate a full system prompt from a short brief
curl --request POST \
  --url https://api.example.com/agents/generate-system-prompt \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "brief": "<string>",
  "name": "<string>",
  "language": "en",
  "company": "<string>",
  "industry": "<string>",
  "tone": "<string>",
  "additional_context": "<string>"
}
'
{
  "system_prompt": "<string>",
  "char_count": 123,
  "model": "<string>",
  "provider": "<string>",
  "tokens_used": 0
}

Documentation Index

Fetch the complete documentation index at: https://docs.mrassistant.ai/llms.txt

Use this file to discover all available pages before exploring further.

Authorizations

​
Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Body

application/json

Brief description the user wants expanded into a full system prompt.

​
brief
string
required

Short description of the agent's purpose, role, and context.

Required string length: 10 - 2000
​
name
string | null

Agent name (used in the prompt's identity section).

Maximum string length: 100
​
language
string | null
default:en

Primary language code, e.g. 'en'.

Maximum string length: 10
​
company
string | null

Company / organization the agent represents.

Maximum string length: 200
​
industry
string | null

Industry or vertical, e.g. 'telecom', 'healthcare'.

Maximum string length: 100
​
tone
string | null

Tone or personality, e.g. 'friendly and professional'.

Maximum string length: 200
​
additional_context
string | null

Optional extra notes — common questions, escalation rules, what to avoid, etc.

Maximum string length: 2000

Response

Successful Response

Generated system prompt ready to drop into agent create/update.

​
system_prompt
string
required

The full generated system prompt.

​
char_count
integer
required

Character count of the generated prompt.

​
model
string
required

Model that produced the prompt.

​
provider
string
required

LLM provider used.

​
tokens_used
integer
default:0

Total tokens consumed for generation.