One call turns a crypto ticker + timeframe into a token-compact market-state brief — so your AI agent never dumps raw OHLCV into context again.
Feeding raw candle arrays to an LLM is token-expensive and error-prone (LLMs hallucinate on numbers). patternfetch returns the digested market state in a fraction of the tokens — patterns, levels, regime and indicator state — plus a one-line summary an agent can act on. Crypto spot, minute-to-daily.
# get a free key
curl -X POST https://patternfetch.com/v1/keys -d '{"email":"you@example.com"}'
# get a brief
curl -X POST https://patternfetch.com/v1/brief \
-H "authorization: Bearer pf_..." \
-d '{"ticker":"BTC/USDT","timeframe":"4h"}'
POST /v1/brief · POST /v1/delta · POST /v1/candles · POST /v1/analogs · POST /v1/keys · GET /v1/platforms · OpenAPI · llms.txt
Tools: patternfetch_brief, patternfetch_delta, patternfetch_analogs, patternfetch_capabilities. Pay per call via x402 (USDC on Base, no account) or Stripe.