How to use a link shortener API inside AI agents and app workflows

Give the agent one narrow tool instead of a broad dashboard abstraction

The best agent tools are clear about input, output, and failure behavior. A shorten route with optional QR support is much easier to reason about than a general-purpose browser automation path.

• The request body is structured and easy to validate before execution.
• The response returns short_url, link id, and QR metadata for downstream steps.
• Stable error codes make it easier to recover from auth, scope, rate, or validation failures.

Wrap the endpoint in one tool function and always send idempotency keys

Agents retry. Workers replay. Tool runners can misfire. The safest default is a dedicated helper that creates a new idempotency key for each new intent and reuses it only on true retry of the same payload.

• Scope the key to the smallest write/read set that flow needs.
• Do not let the agent choose organization ids or dashboard session context.
• Persist request_id values alongside your own run logs when debugging.

export async function createAgentShortLink(input: {
  original_url: string;
  custom_slug?: string;
}) {
  const response = await fetch('https://1get.site/api/v1/links/shorten', {
    method: 'POST',
    headers: {
      Authorization: `Bearer ${process.env.ONEGET_AGENT_KEY}`,
      'Content-Type': 'application/json',
      'Idempotency-Key': crypto.randomUUID(),
    },
    body: JSON.stringify({
      original_url: input.original_url,
      custom_slug: input.custom_slug,
      qr: { include: true, format: 'PNG', size: 256 },
    }),
  });

  const body = await response.json();
  if (!response.ok) {
    throw new Error(`${body.code}: ${body.error}`);
  }

  return body.data;
}

Use audit visibility and usage limits as part of the agent operating model

If an agent can create links, it can create operational noise too. The right answer is not to avoid automation; it is to give automation clear scopes, observable keys, and usage boundaries.

The developer portal and per-key activity history make it easier to understand which automation lane created traffic, when it last ran, and whether the key is still healthy.

• Create separate keys for different agent families or environments.
• Read usage and per-key activity from the dashboard when behavior changes.
• Upgrade into Business or Enterprise when agent traffic is no longer experimental.