# Old English MCP server

> A read-only MCP server and JSON API over the English Wiktionary's Old English
> entries — the West Germanic language of the Anglo-Saxons (Ænglisc, c. 450–1150),
> NOT modern English. Use it to read and write Old English with attested words and
> correct inflected forms instead of inventing them. Read-only.

## What it does
- Looks an Old English word up (any form) and returns its senses plus full
  declension/conjugation tables — the nouns' cases (including the instrumental),
  the adjectives' strong/weak family, the verbs' moods, and the personal
  pronouns' dual number.
- Finds the Old English equivalents of a modern ENGLISH word
  (search_language=eng): each sense's translations from Wiktionary's translation
  tables (the result's translations block), expanded into their full dictionary
  entries.
- Resolves an inflected query to its dictionary lemma automatically via paradigms
  already cached — the result says so (search_method=lemma_index, with
  resolved_from/resolved_lemma). No external analyzer exists for Old English, so a
  form not yet in any cached paradigm won't resolve — look the lemma up directly.
- Folds the special letters, so ASCII spellings work once a paradigm is cached:
  macrons and the dotted ġ/ċ drop, and þ/ð/æ/ƿ fold ('thing' finds þing, 'waeter'
  finds wæter).

## How to connect (MCP)
Remote streamable-HTTP endpoint (no auth, read-only): https://anglo.valksor.com/mcp
  Claude Code:  claude mcp add --transport http anglo https://anglo.valksor.com/mcp
  Cursor / Cline (~/.cursor/mcp.json):
    {"mcpServers": {"anglo": {"url": "https://anglo.valksor.com/mcp"}}}
  stdio-only clients — bridge with mcp-remote:
    {"mcpServers": {"anglo": {"command": "npx",
       "args": ["-y", "mcp-remote", "https://anglo.valksor.com/mcp"]}}}
Tools (each returns Markdown plus typed structuredContent per its outputSchema):
- anglo_search(query, search_language='ang'|'eng', include_forms=true, max_forms=N)
    Senses plus inflection handles (entry_id = the lemma/page title, word_class =
    the part-of-speech section). With include_forms (default TRUE on MCP; the JSON
    API defaults it false), each match's full table comes back inline. With
    search_language='eng' the query is a modern English word and the result lists
    its per-sense Old English equivalents plus their entries.
- anglo_get_inflections(entry_id, word_class)
    The full paradigm for one handle. Its structuredContent nests the paradigm
    under a "result" key (the other tool's payload is unwrapped); the JSON API
    returns it directly. Both shapes are discriminated by "category":
    nominal gives case keys -> {singular, dual, plural} (nouns use the bare
    case; adjectives prefix the strong/weak class and suffix the gender — 'strong
    nominative feminine'; the personal pronouns prefix the person/gender and fill
    the dual); verbal gives finite "<mood> <tense>" -> "<person> <number>" -> form
    plus the non-finite forms (the two infinitives and the participles);
    "not_found" when no table exists.
- anglo_romanize(text)
    Reduces Old English orthography to its primary ASCII spelling (þing→thing,
    wæter→waeter; macrons and the dotted ġ/ċ dropped). Pure transform, no dictionary
    lookup.
- anglo_deromanize(text)
    The inverse: given an ASCII spelling, find the real Old English word(s) whose
    romanization matches. romanize is deeply lossy, so this is a LOOKUP against words
    already cached — attested spellings come back as source=dictionary; on a miss, one
    best-effort reconstruction (source=reconstruction, approximate=true), a guess.
Search tips: one Old English word per call; any form works, and ASCII spellings
('thing', 'waeter') resolve once the lemma's paradigm is cached; for modern
English→Old English pass search_language='eng' (multiword entries like 'apple
tree' work).
Prompts & resources (MCP):
- Prompts anglo_translate_to(text) and anglo_translate_from(text, target_language)
  package the grounded-translation method.
- Resource anglo://glossary/grammar-labels documents the paradigm keys; this
  document is the llms.txt resource.

## How to call (JSON API)
- GET https://anglo.valksor.com/api/search?query=cyning
- GET https://anglo.valksor.com/api/search?query=water&search_language=eng
- GET https://anglo.valksor.com/api/inflections?entry_id=bindan&word_class=verb
- GET https://anglo.valksor.com/api/romanize?text=%C3%BEing   (URL-encoded "þing")
- GET https://anglo.valksor.com/api/deromanize?text=thing   (attested spellings; reconstruction on a miss)
- GET https://anglo.valksor.com/api/health
- Interactive docs: https://anglo.valksor.com/api/docs  ·  schema: https://anglo.valksor.com/api/openapi.json

## Caching and politeness
Dictionary responses are cached server-side with a long TTL (Wiktionary content
changes slowly). Upstream calls are serialised with a polite minimum interval.
Calls are rate limited per client IP — on a 429, wait a few seconds and retry.

## Source and licence
- Dictionary: Wiktionary, the free dictionary (en.wiktionary.org), Old English entries. Wikimedia Foundation. Text available under CC BY-SA 4.0; attribution and share-alike required when republishing.
- Licence: CC BY-SA 4.0 — attribute and share alike when republishing dictionary text.