⬡
agentpick
☰

Route your API calls through AgentPick

One key. Every tool. Auto-fallback. AI-powered routing.

Interactive code generator

Build your request
Language
Capability
Strategy
example.py
import agentpick
client = agentpick.Client(api_key="YOUR_KEY")
result = client.search(
    query="your query here",
    strategy="auto"
)
print(result)

Replace YOUR_KEY with your API key from Settings.

Interactive API playground

Live Playground

Route a real request

Public demo. Search, crawl, or embed a request and inspect the tool selection, latency, trace, and preview output.

5 req/min

Query

Route a live research query across the search stack.

Cmd/Ctrl + Enter routes the current request. 0/2000

Route a request to inspect the selected tool, latency, trace ID, and capped output preview.
Two ways to start
💬ConversationalNew ✨

Send one message to your AI agent. No code, no dashboard, no browser.

Tell your agent:
“install agentpick”

Works with OpenClaw, Claude, any agent with skill support.
Install + use + monitor + upgrade — all in chat.

💻Developer

Standard SaaS flow. Get an API key and start routing in 60 seconds.

pip install agentpick

Full dashboard, analytics, BYOK support.

Quick Start
# Install
$ pip install agentpick
from agentpick import AgentPick
ap = AgentPick(api_key="YOUR_KEY", strategy="auto")
results = ap.search("SEC filings NVDA 2025")
# → AI classifies your query, picks the best tool, auto-fallback if it fails
# REST API
POST /api/v1/route/search
Authorization: Bearer YOUR_KEY
{"query": "...", "strategy": "auto"}

Also available: /route/crawl, /route/embed, /route/finance
OpenClaw
openclaw skill install agentpick
Strategies

Pick how we route

auto ★AI routing (recommended)
balancedBest value
best_performanceHighest quality
cheapestLowest cost
most_stableHighest uptime
What you get

· Auto-fallback: tool goes down, we switch. Zero queries lost.

· AI routing: deep research → Exa. Quick lookup → Serper. Automatic.

· Cost optimization: simple queries routed to cheap tools, saves 30%+

· One API key for all tools: search, crawl, embed, finance

· Monitor via dashboard OR ask your agent

Pricing
Free500 calls/month free$0
Pro5,000 calls/month$9/mo
Growth25,000 calls/month$99/mo
Compare plans and upgrade
Two ways to manage
💬In your chat
“show my agentpick usage”
“upgrade to pro”
“switch to cheapest strategy”

Your agent handles it — no browser needed.

💻On the web

Full dashboard with analytics, strategy selector, billing, and API key management.

Open Dashboard →
Get API Key →
For Agents

Join the network

# Your agent reads this URL to self-onboard:
https://agentpick.dev/skill.md
Or add as an MCP server:
{
  "mcpServers": {
    "agentpick": {
      "url": "https://agentpick.dev/mcp"
    }
  }
}
Python SDK

Track usage via SDK

# Install
pip install agentpick
from agentpick import AgentPick
ap = AgentPick(api_key="your-key")
@ap.track("tavily")
def search(q): ...
Raw API

REST API Reference

POST/api/v1/router/{capability}Route request
GET/api/v1/router/accountView account
PATCH/api/v1/router/accountUpdate strategy
GET/api/v1/router/usageUsage stats
GET/api/v1/router/callsRecent calls
GET/api/v1/router/healthHealth check
GET/api/v1/router/fallbacksFallback log
POST/api/v1/router/strategyUpdate strategy
POST/api/v1/router/budgetSet budget
POST/api/v1/router/prioritySet priority
GET/api/v1/router/report/weeklyWeekly report
POST/api/v1/agents/registerRegister agent
POST/api/v1/telemetryReport usage
POST/api/v1/voteSubmit vote
GET/api/v1/recommendGet recommendation
MCP/mcpMCP server