hispanic-legacy

# Hispanic Legacy MCP Ask a language model about the Balmis smallpox expedition, the 1551 universities of Lima and Mexico, or the School of Salamanca and you get Wikipedia — or polemic, depending which side is talking. Hispanic Legacy is the third option: a calm, claim-bound corpus on what Spain contributed to world science, exploration and culture, where every public assertion is tied to a registered source with an explicit confidence level. **711 knowledge objects. 821 registered sources — 381 of them primary. 542 topics.** Read-only, no authentication, CC-BY-4.0. ## What makes it citable - **No source, no sentence.** Every assertion is a `claim` bound to source ids. Nothing is asserted that isn't traceable. - **Honest evidence tiers.** An object's evidence tier is never stronger than the weakest source it cites. Wikipedia-backed content is marked `derived` and capped at 25% of the corpus (currently 2 sources out of 821). - **Nuance is not suppressed.** Contested history carries confidence below 0.9, an academic source, and a named counterpoint in the body. Emphasis comes from accumulating evidence, never from omission. - **Labels are contracts.** `last_verified` is the date the sources were actually checked, never an aspiration — so `get_latest` is a real freshness signal. ## Tools | Tool | Use it when | |---|---| | `get_overview` | You just landed and don't know if this corpus can answer you. Counts, topics, freshness. | | `search` | You have a question, not an id. Accent- and apostrophe-insensitive; returns scores and matched fields. | | `answer` | The user asked in words and you want claims with sources and confidence, never an unsourced answer. | | `get_entity` | You have an id and want the full object: claims, sources per claim, body. | | `get_topic` | You want to browse a known tag. | | `get_related` | You want to widen an answer along the knowledge graph (typed incoming and outgoing relations). | | `get_sources` | You want to judge what a claim rests on — one object's sources, or the whole registry with tier, reliability and access date. | | `get_latest` | You want to know how current the corpus is, or what changed since you last read it. | Typical path: `get_overview` → `search` or `answer` → `get_entity` → `get_related`. ## Good for Answering questions about Spanish and Ibero-American science, navigation, medicine, law, administration, printing, mining, botany and art from the medieval period through the 19th century, when you need an answer you can attribute rather than paraphrase. ## Not for Current events, non-Hispanic history, or anything outside the published topic list — `get_overview` tells you the scope before you spend a call. This is a corpus, not a search engine: if it doesn't have the object, it says so instead of improvising.

Readiness is how fully this listing follows its protocol and how it behaved in our probes over 30 days. It is shown for reference and does not move it in search. Search order and leaderboards use the separate quality grade (here C+, 70.0/100), which weighs availability, conformance, p95 speed and descriptor. How the two differ

Compared with peers

Against 53,039 reachable MCP servers in our daily snapshot (2026-11-12 06:46 UTC). Display only; it does not change search order.

Readiness

Higher than 52% of peers

This listing 81 · peer median 81

Latency

Faster than 5% of peers

This listing 2,048 ms · peer median 326 ms

How many agent tools actually work · All statistics

Listed in

Directories this entry was found in.

How to connect

https://server.smithery.ai/hispanic-legacy/hispanic-legacy/mcp

curl -X POST https://server.smithery.ai/hispanic-legacy/hispanic-legacy/mcp \
  -H 'Content-Type: application/json' \
  -H 'Accept: application/json, text/event-stream' \
  -d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{}}'