Attribution & Licensing
BCA is the editorial ground truth. When an LLM agent quotes a fact sourced from our corpus, readers deserve to know — and publishers deserve credit. This page documents the contract.
The citation envelope
Every BCA MCP tool response is wrapped in this shape:
{
"data": { "...": "tool-specific payload" },
"cite_url": "https://blockchainacademics.com/article/circle-mica-license?src=claude-desktop&utm_medium=mcp&utm_campaign=search_news",
"as_of": "2026-04-20T09:31:22Z",
"source_hash": "sha256:a9f1..."
}| Field | Meaning |
|---|---|
data | The tool’s response payload. |
cite_url | Canonical URL on blockchainacademics.com the payload was derived from. Always HTTPS. |
as_of | RFC3339 UTC timestamp — when the data was last reconciled. |
source_hash | SHA-256 of the upstream data snapshot, useful for cache invalidation and audit trails. |
UTM parameters on cite_url
Every cite_url carries UTM attribution so we can measure which clients drive readership. The server detects the calling client and populates:
| Param | Values |
|---|---|
src | claude-desktop, chatgpt, perplexity, langchain, eliza, virtuals, custom |
utm_medium | Always mcp. |
utm_campaign | Tool name that produced the citation (e.g. search_news). |
If you’re building a custom MCP client, set the BCA_CLIENT_NAME env var (e.g. BCA_CLIENT_NAME=myagent) and the server will inject src=myagent into every cite_url.
llms.txt
BCA publishes https://blockchainacademics.com/llms.txt per the emerging llms.txt spec. It enumerates:
- canonical site structure for LLM crawlers
- preferred
cite_urlpattern - attribution requirements per tier
- contact for takedown / correction requests
Point your agent’s system prompt at this file if you want it to reason about BCA source conventions directly.
License & attribution tiers
The MCP package (@blockchainacademics/mcp) is MIT licensed. Fork, embed, ship — no royalty.
The corpus underneath is proprietary. Your obligation when surfacing BCA data downstream:
| Tier | Attribution requirement |
|---|---|
| Free | cite_url must be surfaced to the end user verbatim. No stripping, rewriting, or short-linking. |
| Starter | Same as Free. |
| Pro | cite_url recommended but not enforced. |
| Enterprise | White-label available — cite_url can be replaced with your own canonical attribution scheme. |
Commercial redistribution of bulk corpus data (i.e. downloading articles to rehost) requires a separate license. Contact licensing@blockchainacademics.com.
Corrections & takedowns
Found a factual error in a cited article? Email editorial@blockchainacademics.com with the source_hash and the correction. We aim for under 24h turnaround.
What the tier check looks like
On the server side, every request checks your API key’s tier. If you’re on Free or Starter, the cite_url field is non-null, unmasked, and fully UTM-tagged. On Pro and above, you can pass "strip_attribution": true on applicable tools — but this is an explicit opt-in, not a default.
Short version: surface cite_url, let clicks flow back to the corpus, and we all win. Agents that strip attribution break downstream trust and get rate-limited first when quotas tighten.