Card snapshot

This is a frozen copy of the agent's agent-card.json as we observed it at the timestamp above. We capture a new snapshot every time the card's content hash changes. Useful for: forensic drift analysis, verifying downstream callers see the right version, reproducing routing decisions made historically.

{
  "name": "Wikexa Knowledge",
  "description": "Structured encyclopedia, entity and dictionary knowledge for AI agents.",
  "url": "https://wikexa.com",
  "provider": {
    "organization": "Wikexa"
  },
  "version": "1.1.0",
  "capabilities": {
    "streaming": false
  },
  "skills": [
    {
      "id": "lookup",
      "name": "lookup",
      "description": "Look up any entity (person, company, place, concept) and get structured facts plus a ~200-token summary. Resolves aliases and Wikidata Q-ids, so \"Apple\", \"Apple Inc\", and \"Q312\" all return the same entity. Use this instead of fetching a web page: a Wikipedia HTML fetch costs ~15,000 tokens to recover ~500 tokens of facts. Free and unlimited."
    },
    {
      "id": "article",
      "name": "article",
      "description": "Full Wikipedia article as clean structured JSON \u2014 sections as an array, infobox as key/value facts, no HTML or wikitext. Pass `sections` to fetch only the parts you need (e.g. [\"History\"]) and `max_chars` to cap the payload; both exist to keep long articles from flooding your context."
    },
    {
      "id": "define",
      "name": "define",
      "description": "Dictionary lookup from Wiktionary: definitions, senses, part of speech, etymology and pronunciation, grouped by language. One word can be a noun in English and a verb in Latin, and you get both \u2014 pass `language` to narrow it, or `pos` to keep only one part of speech. Use this rather than lookup() for what a word means; lookup() is for entities."
    },
    {
      "id": "search",
      "name": "search",
      "description": "Find entities by fuzzy text match when you do not know the exact title. Returns ranked {title, wikidata_id, description, summary_snippet} \u2014 ranking blends full-text relevance with monthly pageviews, so common meanings win (\"apple\" returns Apple Inc. and the fruit, not disambiguation pages). Follow up with lookup() or article()."
    },
    {
      "id": "recent",
      "name": "recent",
      "description": "Articles changed recently, optionally filtered by topic. This is the freshness escape hatch for facts newer than your training cutoff \u2014 elections, appointments, releases, deaths. Returns titles with timestamps and edit comments; resolve any of them with lookup()."
    }
  ]
}