Agents API | Bootspring Docs

Invoke and manage AI expert agents programmatically.

Endpoints#

Method	Endpoint	Description
`GET`	`/v1/agents`	List available agents
`GET`	`/v1/agents/:name`	Get agent details
`POST`	`/v1/agents/:name/invoke`	Invoke an agent
`GET`	`/v1/agents/invocations`	List invocations
`GET`	`/v1/agents/invocations/:id`	Get invocation result

List Agents#

GET /v1/agents

Query Parameters#

Parameter	Type	Description
`category`	string	Filter by category
`tier`	string	Filter by tier requirement

Example Request#

curl https://api.bootspring.dev/v1/agents \
  -H "Authorization: Bearer bs_xxx"

Response#

{
  "data": [
    {
      "name": "frontend-expert",
      "displayName": "Frontend Expert",
      "description": "React, Next.js, and modern frontend development",
      "category": "development",
      "tier": "free",
      "capabilities": [
        "React component development",
        "Next.js App Router patterns",
        "Tailwind CSS styling",
        "Performance optimization"
      ]
    },
    {
      "name": "backend-expert",
      "displayName": "Backend Expert",
      "description": "APIs, databases, and server-side architecture",
      "category": "development",
      "tier": "free",
      "capabilities": [
        "API design and implementation",
        "Database schema design",
        "Authentication patterns",
        "Server optimization"
      ]
    }
  ]
}

Get Agent Details#

GET /v1/agents/:name

Example Request#

curl https://api.bootspring.dev/v1/agents/frontend-expert \
  -H "Authorization: Bearer bs_xxx"

Response#

{
  "data": {
    "name": "frontend-expert",
    "displayName": "Frontend Expert",
    "description": "Expert in React, Next.js, and modern frontend development",
    "category": "development",
    "tier": "free",
    "capabilities": [
      "React component development",
      "Next.js App Router patterns",
      "Tailwind CSS styling",
      "Performance optimization",
      "Accessibility best practices"
    ],
    "collaboratesWith": [
      "ui-ux-expert",
      "testing-expert",
      "performance-expert"
    ],
    "stats": {
      "invocations": 15420,
      "avgDuration": 2500,
      "satisfaction": 4.8
    }
  }
}

Invoke Agent#

POST /v1/agents/:name/invoke

Request Body#

Field	Type	Required	Description
`prompt`	string	Yes	Task or question
`context`	object	No	Additional context
`options`	object	No	Invocation options

Context Object#

Field	Type	Description
`projectId`	string	Project context
`files`	array	Relevant file paths
`codebase`	string	Codebase summary
`history`	array	Previous messages

Options Object#

Field	Type	Description
`stream`	boolean	Stream response
`maxTokens`	integer	Max response tokens
`temperature`	number	Response creativity (0-1)
`timeout`	integer	Timeout in ms

Example Request#

curl -X POST https://api.bootspring.dev/v1/agents/frontend-expert/invoke \
  -H "Authorization: Bearer bs_xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "Create a responsive pricing table component with three tiers",
    "context": {
      "projectId": "proj_abc123",
      "files": ["components/ui/Card.tsx", "lib/stripe.ts"]
    },
    "options": {
      "maxTokens": 4096
    }
  }'

Response#

{
  "data": {
    "id": "inv_xyz789",
    "agent": "frontend-expert",
    "status": "completed",
    "response": {
      "content": "Here's a responsive pricing table component...\n\n```tsx\n// components/PricingTable.tsx\n'use client';\n\nimport { useState } from 'react';\nimport { Card } from './ui/Card';\n\ninterface PricingTier {\n  name: string;\n  price: number;\n  features: string[];\n  popular?: boolean;\n}\n\nconst tiers: PricingTier[] = [\n  {\n    name: 'Free',\n    price: 0,\n    features: ['1 project', 'Basic agents', 'Community support']\n  },\n  {\n    name: 'Pro',\n    price: 19,\n    features: ['10 projects', 'All agents', 'Priority support'],\n    popular: true\n  },\n  {\n    name: 'Team',\n    price: 49,\n    features: ['50 projects', 'Custom agents', 'SSO', 'SLA']\n  }\n];\n\nexport function PricingTable() {\n  const [annual, setAnnual] = useState(false);\n  // ... component implementation\n}\n```",
      "type": "code",
      "language": "tsx"
    },
    "usage": {
      "inputTokens": 450,
      "outputTokens": 1250,
      "totalTokens": 1700
    },
    "duration": 2340,
    "createdAt": "2024-01-15T10:30:00Z"
  }
}

Streaming Response#

For streaming responses:

curl -X POST https://api.bootspring.dev/v1/agents/frontend-expert/invoke \
  -H "Authorization: Bearer bs_xxx" \
  -H "Content-Type: application/json" \
  -H "Accept: text/event-stream" \
  -d '{
    "prompt": "Create a pricing table",
    "options": {"stream": true}
  }'

Response (Server-Sent Events):

data: {"type":"start","id":"inv_xyz789"}

data: {"type":"content","delta":"Here's a"}

data: {"type":"content","delta":" responsive"}

data: {"type":"content","delta":" pricing table..."}

data: {"type":"done","usage":{"totalTokens":1700}}

List Invocations#

GET /v1/agents/invocations

Query Parameters#

Parameter	Type	Description
`agent`	string	Filter by agent name
`status`	string	Filter by status
`start`	string	Start timestamp
`end`	string	End timestamp
`limit`	integer	Max results

Example Request#

curl "https://api.bootspring.dev/v1/agents/invocations?agent=frontend-expert&limit=10" \
  -H "Authorization: Bearer bs_xxx"

Response#

{
  "data": [
    {
      "id": "inv_xyz789",
      "agent": "frontend-expert",
      "status": "completed",
      "prompt": "Create a pricing table...",
      "duration": 2340,
      "tokens": 1700,
      "createdAt": "2024-01-15T10:30:00Z"
    }
  ],
  "pagination": {
    "hasMore": true,
    "cursor": "cursor_abc"
  }
}

Get Invocation#

GET /v1/agents/invocations/:id

Example Request#

curl https://api.bootspring.dev/v1/agents/invocations/inv_xyz789 \
  -H "Authorization: Bearer bs_xxx"

Response#

Returns full invocation details including the response content.

SDK Usage#

JavaScript/TypeScript#

import { Bootspring } from '@bootspring/sdk';

const bs = new Bootspring({ apiKey: 'bs_xxx' });

// List agents
const agents = await bs.agents.list();

// Get agent details
const frontend = await bs.agents.get('frontend-expert');

// Invoke agent
const result = await bs.agents.invoke('frontend-expert', {
  prompt: 'Create a pricing table component',
  context: {
    projectId: 'proj_123',
  },
});

// Stream response
const stream = await bs.agents.invoke('frontend-expert', {
  prompt: 'Create a pricing table',
  options: { stream: true },
});

for await (const chunk of stream) {
  process.stdout.write(chunk.delta);
}

Python#

from bootspring import Bootspring

bs = Bootspring(api_key='bs_xxx')

# Invoke agent
result = bs.agents.invoke(
    'frontend-expert',
    prompt='Create a pricing table component',
    context={'project_id': 'proj_123'}
)

print(result.response.content)

Agent Categories#

Category	Agents
`development`	frontend-expert, backend-expert, database-expert, api-expert
`quality`	testing-expert, security-expert, performance-expert, code-review-expert
`architecture`	architecture-expert, devops-expert, cloud-expert
`business`	business-strategy-expert, fundraising-expert, growth-expert
`design`	ui-ux-expert, branding-expert

Rate Limits#

Plan	Invocations/Hour	Concurrent
Free	10	1
Pro	100	5
Team	500	20
Enterprise	Custom	Custom

Errors#

Code	Description
`agent_not_found`	Unknown agent name
`invocation_failed`	Agent invocation error
`rate_limit_exceeded`	Too many invocations
`context_too_large`	Context exceeds limits
`timeout`	Invocation timed out