Wikexa Knowledge

Measured conclusion: Deep-tested. L2 outcome: tools_ok.

Wikipedia, Wikidata and Wiktionary as clean JSON, not HTML. 1.9M searchable. Free, no auth.

Add to your agent

Copy this into your MCP client config:

{
  "mcpServers": {
    "Wikexa Knowledge": {
      "url": "https://wikexa.com/mcp",
      "transport": "streamable-http"
    }
  }
}

Measured tools (6)

• lookup

  Facts about any named thing — person, company, place, species, event, concept. Returns structured fields (dates, identifiers, relationships) plus a ~200-token summary, drawn from 10.2M entity records. Prefer this over fetching an encyclopedia page: the HTML costs ~15,000 tokens to recover ~500 tokens of fact. Resolves aliases and Wikidata Q-ids, so "Apple", "Apple Inc" and "Q312" all reach the same entity. Free, no key.

  Parameters:

  • entity (required) — string — Entity name, Wikipedia title, alias, or Wikidata Q-id (e.g. "Tim Cook", "Q312").
  • corpus — string — Which corpus to look in. Defaults to wikipedia. Use wikivoyage for travel guides, wikiquote for quotations, wikibooks for textbooks, wikiversity for course material.

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• article

  The full text of an article, for when lookup()'s summary is not enough — sections as a JSON array, infobox as key/value facts, no HTML or wikitext to parse. Pass `sections` to pull only the parts you need (e.g. ["Early life"]) and `max_chars` to cap the payload; both exist because a long article will otherwise flood your context.

  Parameters:

  • title (required) — string — Article title, alias, or Q-id.
  • corpus — string — Which corpus to read from. Defaults to wikipedia.
  • sections — array — Optional section names to include (substring match, case-insensitive). Omit for the whole article.
  • max_chars — integer — Optional cap on total section text returned.

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• define

  What a word means, in thousands of languages — 8.15M dictionary entries with senses, part of speech, etymology and pronunciation. Covers what a general model is weakest at: historical languages (Old English, Gothic, Ancient Greek, Middle French) and hundreds of regional and indigenous ones. A single spelling often has entries in many languages and you get all of them — `hund` returns Danish, Gothic, Icelandic, Middle English and more — or pass `language` to narrow, `pos` for one part of speech. Use this for words and lookup() for things: define("java") gives the word in eight languages, lookup("Java") gives the island.

  Parameters:

  • word (required) — string — The word or phrase to define.
  • pos — string — Optional part of speech filter, e.g. "Noun", "Verb", "Adjective".
  • language — string — Optional language name as Wiktionary spells it, e.g. "English", "Latin", "Spanish".

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• search

  Find the right title when you only have a partial name or a rough description. Returns ranked {title, wikidata_id, description, summary_snippet}; ranking blends text relevance with monthly pageviews and follows redirects, so abbreviations land on the real article — "usa" returns United States, "jfk" returns John F. Kennedy, "apple" returns Apple Inc. rather than a disambiguation page. Searches every corpus at once unless you pass `corpus`. Follow up with lookup() for facts or article() for the text.

  Parameters:

  • query (required) — string — Free-text search query.
  • limit — integer — Maximum results, 1-50 (default 10).
  • corpus — string — Restrict to one corpus. Omit to search all of them at once, which is usually what you want when you do not know where the answer is.

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• recent

  What changed in the last hours or days — the escape hatch for facts newer than your training cutoff. Reach for this whenever the answer could have moved since you were trained: elections, appointments, acquisitions, releases, deaths, records. Returns titles with timestamps and edit comments; resolve any of them with lookup(). Pass `topic` to filter and `hours` to widen the window up to a week.

  Parameters:

  • hours — integer — Look-back window in hours, 1-168 (default 24).
  • limit — integer — Maximum changes, 1-100 (default 25).
  • topic — string — Optional case-insensitive filter on title or edit comment.

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• papers

  Academic paper metadata from 57M+ works — title, abstract, authors, citations, DOI and open access URL. Covers every field: CS, medicine, physics, economics, biology, and more. Search by keywords (e.g. "transformer attention"), by topic name (e.g. "machine learning"), or by OpenAlex topic ID. Returns papers sorted by citation count. Source: OpenAlex (CC0 metadata). Use this when the user needs scholarly references, citation counts, or research context that Wikipedia does not cover.

  Parameters:

  • year — integer — Publication year to filter on, e.g. 2023.
  • limit — integer — Maximum papers to return, 1-20 (default 5).
  • query — string — Keywords to search for. Matches topic names to find the right research area, then filters by title and abstract. Example: "transformer attention", "CRISPR gene editing".
  • topic — string — Topic name or OpenAlex topic ID. Names are resolved automatically (e.g. "machine learning" → T10135). Use an ID like "T10135" if you already know it. Optional — query alone works.

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

Publisher website

Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.