Enough to ship a prototype. No card and no clock.
- Core read tools: news, market, entities
- 1 year of history and as_of queries for any past date
- Community Q&A on GitHub
- Attribution required on responses
- Rate limit: 2 requests a second
Every plan uses the same endpoint and the same response envelope. Upgrading unlocks more tools and more calls, not a different product. Install the MCP server or call the REST API in under a minute.
All prices in USD, billed monthly through Stripe. Cancel any time; no contract on the self serve plans. The daily cap is what stops a key: past it, or past the per second limit, calls answer HTTP 429 until midnight UTC or the next second. The monthly figure is a guide and is not enforced on its own.
Enough to ship a prototype. No card and no clock.
Enough headroom for a side project agent that runs around the clock.
vs CoinGecko Analyst at $129 a month: 93% less
The plan most agent workloads want: the proprietary indicators, custom indicator formulas, due diligence reports and whitepaper summaries.
vs CoinGecko Lite at $499 a month: 90% less. Messari Pro is $500.
The most calls a self serve plan gets, plus tokenomics models and contract translation, for teams that need the output tomorrow.
vs Kaito Pro at $1,000+ a month: 80% less. CoinGecko Pro is $999.
Negotiated call volume, a contractual SLA and a dedicated engineer. Usually 80% less than Messari Enterprise.
Under Messari Enterprise, typically 80% less
| Plan | Support | Uptime |
|---|---|---|
| Free | Public Q&A on GitHub, email for account issues | Best effort |
| Starter | Email, 48 hour response | Best effort |
| Pro | Email, 24 hour response | 99.9% |
| Team | Private channel on request, 4 hour response | 99.9% |
| Enterprise | Dedicated engineer and Slack Connect | 99.99%, contractual |
Same key, same data and same envelope whichever way you connect.
One command installs all 105 tools. Add the server to your client config with your key.
npx -y @blockchainacademics/mcp{
"mcpServers": {
"blockchainacademics": {
"command": "npx",
"args": ["-y", "@blockchainacademics/mcp"],
"env": { "BCA_API_KEY": "your-key-here" }
}
}
}The Python package wraps the same endpoint. Any HTTP client works too: send your key in the X-API-Key header.
pipx install bca-mcpSpec: openapi.json
19 categories. Every plan reads the same endpoint; higher plans unlock more of them.
Attribution, citations and point in time queries ship on the free plan. The table stakes are not paywalled.
Every response carries cite_url, as_of and source_hash, so an agent's answer has its provenance built in.
Editorial corpus, entity dossiers, live market and on-chain data, grants, events and jobs behind one API key, over REST and MCP.
Articles cited in AI answers are tracked through llms.txt and cite_url, so a publisher sees where its work lands.
as_of queries on every plan, the demo key included: ask for the article corpus as it stood on any past date.
Every plan sees current API health on the status page, checked at most once a minute. Pro and Team target 99.9% uptime; Enterprise sets a contractual figure.
Anything else goes to [email protected], answered within the response time your plan lists above, or to the public Q&A on GitHub Discussions.
Continue with GitHub or Google, or confirm your email, on the signup page. The key is shown once in your browser and never emailed. Paid plans: pay with the Stripe link, then sign in to the dashboard with GitHub, Google or the email you paid with and reveal the key once. If that email already has a free key, it simply moves up to the plan. We store only a one-way hash, so rotate from the dashboard if you lose one.
Free: ask in public on GitHub Discussions, where every answer stays searchable for the next developer; email us only for account issues, best effort. Starter: email, 48 hour response. Pro: email, 24 hour response. Team: a private channel on request (Slack Connect or Telegram, your pick), 4 hour response. Enterprise: a dedicated engineer and Slack Connect. The support address is [email protected]. Include your key prefix in any request, never the full key.
Every plan has a per second limit and a daily cap, counted per account on Free, Starter, Pro and Team. The daily caps are Free 87, Starter 1,083, Pro 10,833, Team 65,000 calls a day (UTC). A call past either limit answers HTTP 429 until the next second or midnight UTC, and the key works again after that. The monthly figure is a guide, not a hard stop, and nothing is billed beyond the plan price. There are no usage emails: the dashboard shows this month's calls. Enterprise limits are negotiated.
Yes, any time and self serve. First cancel your current plan from the billing portal on your dashboard (a free key has nothing to cancel), then pay for the new plan with the email your key uses: the key moves to the new plan as soon as the payment clears. Once the key has moved, the portal shows only the new plan, so if an older plan is still billing, email [email protected] with your key prefix and we cancel it. No contract on Starter, Pro or Team.
No. Starting a job and each poll of it are ordinary calls, counted like any other. Pro runs due diligence reports and whitepaper summaries; Team adds tokenomics models and contract translation. A key can have 3 jobs queued or running at a time, and a fourth answers HTTP 429. Each job returns a job_id you poll.
Every response includes a cite_url pointing at the page the data came from (the article, the upstream source, or the tool's entry in our reference when the data has no page of its own), plus a source_hash for tamper evidence. Free and Starter must surface cite_url in agent answers; Pro and above may keep it internal.
The free plan is the trial: 2,000 calls a month, no card, no expiry. Enough to ship a prototype; upgrade when you outgrow it.
Checkout runs through Stripe: credit and debit cards, Apple Pay, Google Pay and Link. Invoices are available on request for the Team plan.
Not yet. It is on the roadmap, USDC on Base for Team and Enterprise first. Email [email protected] with the subject "crypto billing" to join the waitlist.
2,000 calls a month with no card and no time limit. Above Team, or for volume, data residency and a contractual Enterprise SLA, talk to us.