Fragmento de Prompt de Sistema

Agrega l402-kit al prompt de sistema de tu agente para que pague por APIs de forma autónoma.

Fragmento para copiar y pegar

Agrega esto al prompt de sistema de cualquier agente para habilitar pagos L402 autónomos:

You have access to paid APIs via the L402 protocol.
When you call an API that returns HTTP 402, use the L402Client from l402-kit
to pay automatically with Bitcoin Lightning.
Budget limit: {BUDGET_SATS} sats per session. Never exceed this limit.
Report spending at the end of each session.

Reemplaza {BUDGET_SATS} con el presupuesto de tu sesión (por ejemplo, 1000 para ~$0.50).

---

LangChain

from langchain.agents import AgentExecutor
from l402kit import L402Client
from l402kit.langchain import L402Tool

wallet = BlinkWallet(api_key=os.environ["BLINK_API_KEY"])
client = L402Client(wallet=wallet, budget_sats=1000)

tools = [L402Tool(client=client, name="paid_api")]

system = """You have access to paid APIs via the L402 protocol.
Budget: 1000 sats per session. Use the paid_api tool when you need data."""

agent = AgentExecutor(agent=..., tools=tools, system_message=system)

---

OpenAI Agents SDK

from agents import Agent, Runner
from l402kit import L402Client
from l402kit.wallets import BlinkWallet

wallet = BlinkWallet(api_key=os.environ["BLINK_API_KEY"])
client = L402Client(wallet=wallet, budget_sats=1000)

agent = Agent(
    name="DataFetcher",
    instructions="""You have access to paid APIs via the L402 protocol.
    When an API returns 402, call pay_and_fetch() with the URL.
    Budget: 1000 sats per session.""",
    tools=[client.as_openai_tool("pay_and_fetch")]
)

---

CrewAI

from crewai import Agent, Task, Crew
from l402kit import L402Client

client = L402Client(wallet=wallet, budget_sats=500)

researcher = Agent(
    role="API Researcher",
    goal="Fetch data from paid APIs autonomously",
    backstory="""You can access paid APIs using Bitcoin Lightning micropayments.
    When you encounter a 402 response, pay with the l402_fetch tool.
    Budget: 500 sats.""",
    tools=[client.as_crewai_tool()]
)

---

Vercel AI SDK

import { l402Client } from "l402-kit/agent";
import { tool } from "ai";

const client = l402Client({ wallet: blinkWallet, budgetSats: 1000 });

const systemPrompt = `You have access to paid APIs via the L402 protocol.
When you need data from a paid source, use the fetchPaid tool.
Budget: 1000 sats per session.`;

const tools = {
  fetchPaid: tool({
    description: "Fetch data from an L402-protected API",
    parameters: z.object({ url: z.string() }),
    execute: async ({ url }) => client.fetch(url),
  }),
};

---

Consejos

• Configura budgetSats de forma conservadora — los agentes pueden acumular llamadas rápidamente
• Usa spendingReport() para registrar el uso al final de la sesión
• Para producción, usa BudgetManager con límites por usuario