Guide

Limits

Page sizes, plan history windows and request rates for the API and the MCP server.

Page sizes

WhereDefaultMaximum
REST list endpoints20100. A larger limit is capped, not rejected
GET /v1/events20200
search_events1550
search_people, search_companies20100
resolve_search_terms2050
get_company_aum_history24120

Plan history windows

The Starter plan includes the last 30 days of events. Higher plans include the full history. Profiles and organisations aren't windowed on any plan.

search_events states the window in every response under applied.date_window, so an empty result is never mistaken for a quiet period.

Request rates

Chikara Intelligence doesn't apply a per-minute request limit to the API or the MCP server at present. Per-plan limits will be published on this page before they're enforced.

Build for them now. Treat a 429 as a signal to wait and retry with exponential backoff, and keep concurrent requests to a handful per token.

Retry on 429
async function get(url, attempt = 0) {
  const res = await fetch(url, {
    headers: { Authorization: `Bearer ${process.env.CHIKARA_API_TOKEN}` }
  });
  if (res.status === 429 && attempt < 5) {
    await new Promise((r) => setTimeout(r, 2 ** attempt * 500));
    return get(url, attempt + 1);
  }
  if (!res.ok) throw new Error(`${res.status} ${await res.text()}`);
  return res.json();
}

const { success } = await get("https://api.chikaraintel.com/v1/events?eventType=move&limit=20");
console.log(success.length);

Fair use

The Terms cover acceptable use, including not working around published limits. If a workload needs more than a handful of concurrent requests, book a call and we'll size it with you.

Related