Rate Limits & Errors

Understanding rate limits, response headers, and error formats.

Rate Limit Tiers

Three windows per plan: per-minute (abuse cap, all plans), per-day (rolling 24h, free only), and per-period (your billing window). See Pricing for prices.

Plan	Per Minute	Per Day (rolling 24h)	Per Period	Period
Free	10	20	600	rolling 30 days
Lawyer Professional	300	—	5,000 per seat	subscription-anchored month
Lawyer Professional Plus	600	—	25,000 per seat	subscription-anchored month
Developer Pro	600	—	100,000	subscription-anchored month
Developer Scale	1,000	—	1,000,000	subscription-anchored month
Commercial & Enterprise	Contract	—	Contract	Contract

Included paid allowances reset monthly on the subscription anchor, including annual subscriptions. Lawyer team allowances are pooled and multiply by purchased seats. The free tier uses a rolling 30-day window over your audit log. MCP protocol handshake (initialize, tools/list) does not count; only actual tool calls do.

Rate Limit Headers

Every response includes headers so you can track your usage in real time. Headers use lowercase names per HTTP/2.

Header	Description
`x-ratelimit-limit-minute`	Requests allowed per minute (abuse cap, all plans)
`x-ratelimit-remaining-minute`	Requests remaining this minute
`x-ratelimit-limit-day`	Rolling 24h cap (free plan only)
`x-ratelimit-remaining-day`	Requests remaining in the last 24h window
`x-ratelimit-limit-period`	Period quota (Stripe billing window for paid; rolling 30d for free)
`x-ratelimit-remaining-period`	Requests remaining in the current period
`x-ratelimit-period-type`	`stripe` (paid) or `rolling_30d` (free)
`x-ratelimit-reset-period`	ISO-8601 reset timestamp (paid only; omitted for rolling windows)
`Retry-After`	Seconds to wait (only on 429 responses)

429 Too Many Requests

When you exceed a quota the API returns a 429 status with a structured error body.

{
  "type": "error",
  "error": {
    "type": "rate_limit_error",
    "message": "Plan quota exceeded for current billing period (free plan). Resets as old requests age out. Upgrade: https://legaldatahunter.com/pricing",
    "limit_type": "period",
    "plan": "free",
    "upgrade_url": "/pricing",
    "period_resets_at": null
  }
}

The limit_type field indicates which quota was exceeded: minute, day, or period. MCP requests receive the same payload wrapped in a JSON-RPC 2.0 error with code -32029.

Exempt Endpoints

The following endpoints are not rate-limited and do not count against your quotas:

GET /v1/discover/* — Discovery endpoints
GET /v1/stats — Statistics
GET /v1/manifest — Manifest
GET /v1/billing/* — Billing endpoints
GET /v1/usage/* — Usage endpoints
GET /v1/keys — Key management
GET /auth/* — Authentication

HTTP Status Codes

Code	Meaning
`200`	Success
`400`	Bad request — invalid parameters
`401`	Unauthorized — missing or invalid authentication
`404`	Not found — document or resource doesn't exist
`429`	Rate limit exceeded
`500`	Internal server error

Error Response Format

All errors follow a standard format with a detail field containing a human-readable message.

General errors

{
  "detail": "Human-readable error message"
}

Validation errors (400)

{
  "detail": [
    {
      "loc": ["body", "q"],
      "msg": "field required",
      "type": "value_error.missing"
    }
  ]
}