Developers / Public API

The Kresmion API

Market intelligence for bots and apps. One HTTP API over prediction markets, on-chain crypto flows, equities, and macro, with a cited source behind every record. Authenticate with a Bearer key, read the live OpenAPI spec, subscribe to webhooks, and pull bulk CSV for backfills. It is information-only and does not provide investment advice.

Quickstart

Mint a key at /settings/api-keys, then fetch five prediction markets. Send the key as a Bearer token on every request.

curl -s "https://kresmion.com/api/v1/prediction/markets?limit=5" \
  -H "Authorization: Bearer krm_your_key_here"

Products

Six product areas, each with representative endpoints. The full, authoritative list is in the live reference below.

API reference

Rendered live from the OpenAPI spec: every endpoint grouped by tag, with parameters and responses. Expand a row for details.

Rate limits and tiers

Every registered account can self-serve an API key. Requests over the per-minute limit get a 429 with a Retry-After header telling you how long to wait.

Need higher limits or a dedicated tier? Email solal@kresmion.com.

Webhooks

Subscribe to push events instead of polling. Manage up to five webhooks per account at /settings/webhooks. Each delivery is signed so you can verify it came from Kresmion.

{
  "id": "evt_01J8XM2QF3",
  "event": "whale.transfer",
  "created_at": "2026-07-16T14:03:22Z",
  "data": {
    "asset": "ETH",
    "amount_usd": 8250000,
    "direction": "to_exchange",
    "counterparty": "Binance",
    "chain": "ethereum"
  }
}

import hashlib
import hmac

def verify(secret: str, raw_body: bytes, signature: str) -> bool:
    expected = hmac.new(secret.encode(), raw_body, hashlib.sha256).hexdigest()
    return hmac.compare_digest(expected, signature)

# In your handler, read the raw request body and the signature header:
#   sig = request.headers["X-Kresmion-Signature"]
#   if not verify(WEBHOOK_SECRET, raw_body, sig):
#       return 401  # reject a forged payload

Bulk data

For backfills and research, pull whole datasets as gzip CSV, ranged by from and to dates. Stream them straight to a file.

curl -s "https://kresmion.com/api/v1/bulk/whales?from=2026-06-01&to=2026-06-30" \
  -H "Authorization: Bearer krm_your_key_here" \
  --output whales_2026_06.csv.gz

MCP for AI agents

AI agents connect to Kresmion over the Model Context Protocol. The hosted server exposes 14 tools covering prediction markets, on-chain flows, ETF flows, derivatives, equity signals, 13F holdings, and the macro regime. Every tool response carries the same envelope and freshness stamp as the HTTP API, and the same key, quota, and rate limits apply.

Hosted endpoint (streamable HTTP), for Claude Desktop and other MCP clients:

{
  "mcpServers": {
    "kresmion": {
      "type": "http",
      "url": "https://kresmion.com/api/mcp",
      "headers": { "Authorization": "Bearer krm_your_key_here" }
    }
  }
}

One line in Claude Code:

claude mcp add --transport http kresmion https://kresmion.com/api/mcp \
  --header "Authorization: Bearer krm_your_key_here"

Prefer to run it locally? Download the standalone stdio server and set your key in the environment:

# download the standalone server, then register it over stdio
curl -sO https://kresmion.com/sdk/kresmion_mcp.py
pip install mcp httpx

{
  "mcpServers": {
    "kresmion": {
      "command": "python",
      "args": ["kresmion_mcp.py"],
      "env": { "KRESMION_API_KEY": "krm_your_key_here" }
    }
  }
}