io.github.Embassy-of-the-Free-Mind/sourcelibrary

Est-ce que io.github.Embassy-of-the-Free-Mind/sourcelibrary est sûr ?

Au dernier audit MCP Hub, le serveur MCP « io.github.Embassy-of-the-Free-Mind/sourcelibrary » est classé « Non encore testé ». Référencé dans l'annuaire. Seul le pentest comportemental fait foi, et il n'a pas encore été exécuté sur ce serveur. Le verdict est réévalué en continu ; consultez la fiche pour l'état à jour.

Méthode : MCP Hub analyse la description de chaque outil (couche 1) et, pour le verdict fort, exécute le serveur dans une micro-VM isolée avec leurres et réseau bloqué (couche 2). On publie des noms et des compteurs, jamais le contenu brut renvoyé par un outil.

Les 22 outils

• check_duplicate≈ 487 tokens

  DOES THE LIBRARY HOLD THIS WORK? Call this BEFORE telling the user the library lacks a book, and before proposing one for acquisition. A miss from list_books or search_library is NOT evidence of absence: list_books matches words and search_library ranks by relevance, so a held book can sit behind dozens of better-ranked hits. Give a title + author (+ year if known), or a library URL (Internet Archive, Gallica, e-rara, BSB, a IIIF manifest). Matches the same scan, the same edition, other editions and the same work, plus near and cross-lingual titles (title in any language). Returns a verdict (same_object / same_edition / possible_same_edition / other_edition / related_title / new), confidence, matches with reasons, ids and URLs, held_not_public (copies held but not yet published: counted, not described), and limits (what the check could not search). A "new" verdict from a URL alone has not searched other editions; pass the title and author too.

  Paramètres, retour & exemple (6)

  Paramètres

  • url · string · optionnel
    A library URL or IIIF manifest for a specific copy (title then optional).
  • year · number · optionnel
    Edition year, if known. Separates "same edition" from "another edition".
  • ia_id · string · optionnel
    Internet Archive identifier, for an exact same-scan match.
  • title · string · optionnel
    Title in any language, original or English, e.g. "Musurgia universalis".
  • author · string · optionnel
    Author in any form ("Athanasius Kircher", "Kircher, Athanasius"). Strongly improves matching.
  • language · string · optionnel
    Language hint for the cross-lingual title match (optional).

  Exemple d'appel

  {"name": "check_duplicate", "arguments": {"url": "texte", "year": 1, "ia_id": "texte", "title": "texte", "author": "texte", "language": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• continue_review_shift≈ 232 tokens

  Next page of the book being reviewed. Call it straight after finishing a page until the stretch the user chose is done, and again when they ask for more. Pass next.book and next.after from the previous page, and volunteer_code. Returns the same shape as start_review_shift; follow how_to_run again. When page is null the book is finished: say so and offer start_review_shift with new_book true.

  Paramètres, retour & exemple (3)

  Paramètres

  • book · string · requis
    next.book from the previous page
  • after · number · requis
    next.after from the previous page
  • volunteer_code · string · requis

  Exemple d'appel

  {"name": "continue_review_shift", "arguments": {"book": "texte", "after": 1, "volunteer_code": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• get_book≈ 730 tokens

  READ PIPELINE step 1 — DISCOVER. START HERE for any named work or author. Returns the book's AI-generated summary, chapter list, edition metadata, DOI, page counts, IIIF manifest, and the cover image (inline, so you and the user can see the book). LANGUAGE: `language` is what is printed on THIS EDITION's leaves, which is frequently not the language the work was written in. Where the record knows they differ the response also carries `work_language` and a `translation_note`, and `text_role` (original / period-translation / modern-translation / unclassified) says what kind of witness this is — read them before describing a passage as the author's own words, because an edition can be a translation of a translation (de Slane's 1863 French Muqaddimah, read in English, is English←French←Arabic). Absent `work_language` means the work's language is NOT RECORDED on this record. It is not a statement that the edition is in the work's own language (most records have never been checked; Risner's 1572 Latin Alhazen carries none, and the work is Arabic). Only `text_role: "original"` asserts that these are the author's own words in the author's language; `text_role: "unclassified"` means nobody has examined it, so do not infer the work's language from `language`, and do not conclude the library lacks a tradition because a `language` filter came back empty. Use list_editions to find an original-language witness. The summary is typically a multi-paragraph orientation covering the book's argument, structure, and significance — often answering the question without further searching. Then: get_book_text to read a chapter or page range (step 2), get_quote / get_quotes to lock specific pages with full citation apparatus (step 3). search_within_book locates passages inside this book. MULTI-WORK VOLUMES: where the scans carry running heads, contains_works lists the works the volume ACTUALLY holds with their page spans, taken from the heads the printer put on each leaf. Trust it over the title — collected-works titles routinely name works the volume does not contain, and the volume holding a work often does not name it. If contains_works is absent the scans have no heads to read; status "insufficient-heads" means it was examined and could not be decided. EDITIONS: `editions` says which languages this book can be READ in and how many pages each covers ({ en: 357 } on most books, { en: 357, es: 357 } where a localized edition exists). Pass a code as `lang` to get_quote, get_quotes, get_book_text or search_within_book to read that edition; without it you get English.

  Paramètres, retour & exemple (1)

  Paramètres

  • book_id · string · requis
    The book ID

  Exemple d'appel

  {"name": "get_book", "arguments": {"book_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• get_book_text≈ 701 tokens

  READ PIPELINE step 2 — READ. Read a book's text. Call get_book first (step 1) for the chapter list, then come here. Preferred: use the chapter param to read one chapter at a time (includes [Page N] markers for citation). Alternatively, use from/to for explicit page ranges (e.g. from=1 to=50). When you find passages worth quoting, hand the page numbers to get_quote / get_quotes (step 3) for verbatim text + a citation link. TRUNCATION: the response always includes truncated: true/false. When truncated=true, the truncation_note field gives the exact next from/to values to call — this means content was cut short by a page-budget limit, NOT that the book ended. An AI agent MUST NOT infer end-of-book from pages_returned alone; check truncated first. Daily page budgets apply across get_book_text/get_quote/get_quotes: anonymous 500 pages/24h, signed-in 1,000, free Explorer keys 2,000, paid keys uncapped — sign in at sourcelibrary.org/auth/signin or get a key at sourcelibrary.org/developers. Corpus-scale text belongs on the dataset API (sourcelibrary.org/dataset), not on this tool.

  Paramètres, retour & exemple (8)

  Paramètres

  • to · number · optionnel
    End page number (inclusive). Recommended chunk size: 50 pages. If the response has truncated=true, use the next from/to from truncation_note.
  • from · number · optionnel
    Start page number (inclusive). Use with to for explicit page ranges.
  • lang · string · optionnel
    ISO code of the EDITION to read, e.g. "es". Default "en". Resolved PER PAGE: a page with no text in that language comes back as English, labelled `translation_lang: "en"` (json) or `[Translation — en]` (plain), and the response carries lang_coverage. Chapter text is materialized in English only, so `chapter` with a non-English `lang` is served from the chapter's page range instead.
  • part · number · optionnel
    Part number (1-based) for large chapters split into multiple parts
  • format · string · optionnel · json | plain
    json (default, structured with per-page fields) or plain (concatenated text with page markers)
  • book_id · string · requis
    The book ID
  • chapter · number · optionnel
    Chapter index (0-based). Preferred over from/to — returns pre-structured chapter text with embedded [Page N] markers.
  • content · string · optionnel · ocr | translation | both
    Which text to include: ocr (original language), translation (English), or both (default)

  Exemple d'appel

  {"name": "get_book_text", "arguments": {"book_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• get_locus≈ 773 tokens

  Turn a CANONICAL CITATION into the actual leaves that carry it. Aristotle is cited by Bekker number (1094a8, 1447a) and Plato by Stephanus number (Rep. 328b) — the references scholarship has used for centuries, which survive re-typesetting and are shareable in a way a scan page never is. USE THIS FIRST whenever a passage arrives as a canonical reference rather than a page: do not try to derive the page yourself from a book's pagination, which is what produced a wrong guess before this tool existed. Bekker numbers are unique across the whole Aristotelian corpus, so the number alone is enough and it also tells you WHICH WORK you are citing. Stephanus numbers restart in each of the three 1578 volumes, so pass work ("Republic", "Timaeus") — without it the response lists the candidate dialogues instead of choosing one. Returns every witness the library holds: the Greek reference edition and, where we have one, an English translation of the same lines, each with its scan page, a reader URL and a quote_api link — so you can compare the original against a translation at one reference. Then call get_quote with the returned book_id + page for the verbatim text and a citable shortlink. LIMITS, stated plainly: a witness is only returned where the reference is PRINTED on that leaf (or, in the two root editions, where a verified constant offset brackets it) — nothing is interpolated, so an empty result means this library holds no anchored leaf there, NOT that the citation is wrong; editions_searched shows what was consulted and the range each covers. Line numbers (the "8" of 1094a8) are not resolved — you get the right leaf and read the line off it. Two works can share a page where one ends and the next begins (Bekker 184 and 1447 are both such joins), and each leaf is filed by the running head printed on it, so a reference at the very start of a work may come back under its predecessor — always read other_works_at_this_reference before concluding a passage is absent. A bare number that exists in both systems returns Aristotle and Plato leaves together; check the system field on each.

  Paramètres, retour & exemple (3)

  Paramètres

  • work · string · optionnel
    The work or dialogue, when the reference needs it (Plato always does): "Republic", "Timaeus", "Laws", "Nicomachean Ethics", "Poetics". Greek or Latin titles as printed in the editions also resolve.
  • system · string · optionnel
    Optional: "bekker" or "stephanus". Inferred from the work when omitted; do not guess it from the number, since the two ranges overlap.
  • reference · string · requis
    The canonical reference: "1094a8", "1094a", "1447", "328b". A leading system name is accepted ("Bekker 1094a"), as is a work name ("Rep. 328b").

  Exemple d'appel

  {"name": "get_locus", "arguments": {"reference": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• get_quote≈ 998 tokens

  READ PIPELINE step 3 — CITE. Get the exact verbatim text of a single page plus its citation apparatus. ALWAYS use before putting text in quotation marks. The response headline is citation_link (the stable sourcelibrary.org/q/… shortlink) — present it to the user alongside the quote. Render as: > [exact translation text, verbatim] > — [Author], [citation.locator]. [citation_link] PAGE BREAKS: this corpus is paginated from physical leaves, and nearly one prose page-boundary in five has a sentence running across it — sometimes a word split by a hyphen ("…our move-" / "movements…"). A page that opens or breaks off mid-sentence still reads as complete prose and still carries a perfectly valid citation, so check the continuity field on every response BEFORE quoting: if continues_on_next or continues_from_previous is true, call again with context: true and quote the whole sentence. Quoting a fragment as though it were the author's complete thought is a misattribution even when the page number is right. NON-LATIN SCRIPTS: where the page is Greek, Hebrew, Arabic, Sanskrit, Cyrillic and so on, the response also carries romanized — the romanization of the original — so the citation can be shown in three layers: original → romanized → translation → citation_link. It is AI-generated reading apparatus, not a transcription; quote the source from original or translation, never from romanized. Absent on Latin-script pages and on non-Latin pages not yet romanized. ENGLISH ORIGINALS: where the leaf is already English there is no translation and none is needed — the response omits `translation`, sets `text_source: "ocr_original"`, and the verbatim text is `original` (with a `transcription_note`). Quote it as the source's own words, never as a translation, and expect period spelling and long-s (ſ) — it is an uncorrected transcription of the scan. `text_source` is on every response (`translation` otherwise), so branch on it rather than guessing from pages_translated, which is 0 for an English-original book by construction. TRANSLATED EDITIONS: `original` means the text printed on this leaf, which on a translated edition is the TRANSLATOR's language, not the author's. When the response carries `translation_note`, the chain is stated there — attribute the wording to the translator and do not offer the passage as evidence of what the author wrote in their own tongue. Call list_editions to find an original-language witness of the same work. For several pages of one book at once, use get_quotes.

  Paramètres, retour & exemple (5)

  Paramètres

  • lang · string · optionnel
    ISO code of the EDITION to quote, e.g. "es". Default "en". `quote.lang` on every response says which edition was actually served — where no such edition exists the English translation comes back with lang: "en" and a lang_note, and it must not be presented as the requested edition. The citation link follows the edition served.
  • page · number · requis
    Page number
  • book_id · string · requis
    The book ID
  • context · boolean · optionnel
    Also return the full text of the previous and next pages, so a sentence spanning the page break can be read whole. Set this when continuity.continues_from_previous or continues_on_next came back true on an earlier call, or whenever you are about to quote near a page edge.
  • include_image · boolean · optionnel
    Also return the scan of the cited leaf as an inline image (display size, ≤1200px). Set this when the user would benefit from SEEING the page — an illustrated leaf, a title page, a diagram, disputed OCR — or asks to see it. The image arrives as an MCP image block you can view and the user sees rendered.

  Exemple d'appel

  {"name": "get_quote", "arguments": {"page": 1, "book_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• get_quotes≈ 538 tokens

  READ PIPELINE step 3 — CITE, in batch. Get verbatim text + citation_link for SEVERAL pages of a single book in one round-trip, to assemble a multi-passage dossier. Specify either pages (an explicit array, e.g. [12, 40, 41]) or an inclusive from/to range. Max 25 pages per call. Each entry carries its own citation_link to present alongside the quote, and — on non-Latin-script pages that have one — a romanized layer to show between the original and the translation (AI apparatus, not a transcription). Every entry also carries `text_source`: `translation` normally, or `ocr_original` on a leaf that is already English, where the verbatim text is `original` and must be attributed as the source's own words rather than as a translation. One batch can mix both — a Latin volume can hold an English preface. Batch pages count toward the shared daily page budget (see get_book_text); this is a citation tool, and corpus-scale extraction belongs on the dataset API (sourcelibrary.org/dataset).

  Paramètres, retour & exemple (6)

  Paramètres

  • to · number · optionnel
    End page (inclusive) of a range. Use with from.
  • from · number · optionnel
    Start page (inclusive) of a range. Use with to.
  • lang · string · optionnel
    ISO code of the EDITION to quote, e.g. "es". Default "en". Resolved per page — one batch can mix editions, so read `quote.lang` on each entry rather than assuming the whole dossier is in one language.
  • pages · array · optionnel
    Explicit list of page numbers (e.g. [12, 40, 41]). Use this OR from/to.
  • book_id · string · requis
    The book ID
  • include_image · boolean · optionnel
    Also return page scans as inline images (display size). The first 5 pages of the batch get inline image blocks; every entry still carries its page_image_url in the JSON.

  Exemple d'appel

  {"name": "get_quotes", "arguments": {"book_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• list_books≈ 772 tokens

  BROWSES/FILTERS THE CATALOG by metadata (canonical author via author_id, title/author/summary words via search, language, category, year range, translation recency) — no page-text matching. PICK THIS to see WHAT EXISTS by an author or in a tradition, or to enumerate a date range. Returns books with title, author (string + canonical author_id + author_url), language, year, and translation progress. → For a relevance-ranked topic search use search_library; for passages on a theme use search_translations (exact words) or search_concept (by meaning).

  Paramètres, retour & exemple (11)

  Paramètres

  • sort · string · optionnel · recent-translation | recent | title-asc | title-desc
  • limit · number · optionnel
    Max results (default 100, max 200)
  • search · string · optionnel
    Relevance-ranked word match over title (original, display and English), author, original-script name forms (e.g. Greek), AND the book's AI summary. A hit can name your term in neither title nor author, only in what the book discusses. Diacritic-insensitive. For exactly one person's books use author_id instead.
  • work_id · string · optionnel
    Every witness of one WORK, across editions, languages and centuries — pass the work_id from a result row or from get_book. Use this instead of a title search whenever you already hold the identifier: witnesses of one work are often titled differently in each language, so a title search finds only the subset that happens to share your wording. Unlike list_editions (which returns the whole set), this composes with language, year_from/year_to and sort.
  • year_to · number · optionnel
    Latest edition year (inclusive).
  • category · string · optionnel
  • language · string · optionnel
  • author_id · string · optionnel
    Canonical author slug, e.g. "jan-hus" — filters to exactly that person's books via the author thesaurus (variant slugs resolve to the canonical person). Every book row in results carries its author_id; the response echoes the canonicalized author.
  • year_from · number · optionnel
    Earliest edition year (inclusive). Matches only books with a known numeric year — ~60% of the library.
  • edition_key · string · optionnel
    Other digitizations of ONE printing — pass the edition_key from a result row. Only full-quality keys match on both sides, so this answers "what other scans of this exact edition do you hold?" and never merges different printings of the same title.
  • has_edition · string · optionnel
    ISO code — return only books READABLE in that language, e.g. "es". Different from `language`, which is the language printed on the leaves of the scan: a Latin book with a Spanish edition matches language="Latin" AND has_edition="es". Each result then also carries url_localized.

  Exemple d'appel

  {"name": "list_books", "arguments": {"sort": "recent-translation", "limit": 1, "search": "texte", "work_id": "texte", "year_to": 1, "category": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• list_editions≈ 306 tokens

  Every edition of one work that the library holds — the other witnesses to the same text, across languages and centuries. Give it a book_id (easiest: the id of any edition you already found) or a work_id. USE THIS when a quotation needs checking against more than one witness, when you want the original-language text behind a translation, or when comparing how a passage reads across editions — differences between witnesses are often the finding. Returns language, date, page count and translation coverage per edition, so you can pick the right one to read. Note: for multi-volume collected works the identifier names the SET rather than a single text, and the response says so explicitly when that applies.

  Paramètres, retour & exemple (2)

  Paramètres

  • book_id · string · optionnel
    Any edition you already have. Its work is looked up and the siblings returned.
  • work_id · string · optionnel
    A work identifier, if you already have one (from get_book).

  Exemple d'appel

  {"name": "list_editions", "arguments": {"book_id": "texte", "work_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• list_image_facets≈ 390 tokens

  List the legal values of search_images' type, subject, figure and symbol filters, with how many images carry each. Those filters are EXACT string matches on tags an AI extractor wrote, and each vocabulary runs to thousands of strings (e.g. "musical notes", "Jesus Christ", "medicinal plants"), so a guessed value usually matches nothing. Look it up here first. Call with no facet for the top values of all four; with facet + contains to find the exact spelling of a term ("ouro" → "ouroboros"). Counts cover book illustrations of visible books only (the museum-artwork collection is not counted), and only strings on 3 or more images are listed; the list is refreshed weekly. If contains finds nothing, the tag is absent or rare: use it as free text in search_images' query instead. No URL in this result is a browse page; link images from search_images results.

  Paramètres, retour & exemple (3)

  Paramètres

  • facet · string · optionnel · type | subject | figure | symbol
    Which filter's vocabulary. Omit for the top values of all four.
  • limit · number · optionnel
    Values per facet (default 100 with a facet, 25 without; max 500).
  • contains · string · optionnel
    Case-insensitive substring to look for, e.g. "dragon". Matches are ordered by image count.

  Exemple d'appel

  {"name": "list_image_facets", "arguments": {"facet": "type", "limit": 1, "contains": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• propose_collection≈ 362 tokens

  Propose a themed collection of books to the Source Library team — a title, a rationale (why these books belong together and what thread connects them), and an ordered list of book ids. Get book ids from search_library / list_books / get_book. Like submit_feedback and share_findings, this goes to the team for REVIEW — it does NOT create a public collection instantly; a curator reviews and approves it. Use this when the user has identified a coherent set of books worth grouping and wants to contribute that curation back.

  Paramètres, retour & exemple (6)

  Paramètres

  • name · string · optionnel
  • email · string · optionnel
  • title · string · requis
    Title of the proposed collection (2-200 chars)
  • book_ids · array · requis
    Ordered list of book ids to include (1-200). Get ids from search_library / list_books / get_book.
  • rationale · string · requis
    Why these books belong together and what connects them (max 5000 chars)
  • suggested_slug · string · optionnel
    Optional URL slug suggestion, e.g. "renaissance-astronomy"

  Exemple d'appel

  {"name": "propose_collection", "arguments": {"title": "texte", "book_ids": [], "rationale": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• propose_correction≈ 688 tokens

  During a review shift, propose a correction to a page's transcription (field "ocr") or English (field "translation"). A correction is a list of exact span edits: each "find" must be copied exactly from the text the shift returned and occur once on the page (add surrounding words if it repeats); "replace" is the corrected text ("" deletes). Fix only what the scan or the original shows is wrong: misread letters, dropped or added words, a sentence the English skips. Do not restyle or modernise. Propose transcription fixes before translation fixes, as separate calls. Pass base_hash = page.text_version for that field. A proposal does NOT change the page: it is applied only after a second reader agrees, and the volunteer is credited. If the call says the page changed since it was loaded, say so and move on. drafted_by: "assistant" if you found and worded it in a book run, "volunteer" if the user supplied the wording, "assistant_accepted" if you drafted it and the user approved it specifically, "mixed" otherwise.

  Paramètres, retour & exemple (10)

  Paramètres

  • name · string · optionnel
    How the user wants to be credited (optional)
  • note · string · optionnel
    Anything the second reader should know (optional)
  • edits · array · requis
    Span edits (1-30).
  • field · string · requis · ocr | translation
    "ocr" for the transcription, "translation" for the English
  • item_id · string · requis
    page.item_id from the shift
  • shift_id · string · optionnel
  • base_hash · string · requis
    text_version.ocr or text_version.translation from the page
  • drafted_by · string · requis · assistant | volunteer | assistant_accepted | mixed
  • volunteer_code · string · requis
  • assistant_model · string · requis
    Your own model name and version, copied exactly as your instructions or system prompt give it, and whether extended thinking is on. Write "unknown" rather than guess; a wrong name is worse than none.

  Exemple d'appel

  {"name": "propose_correction", "arguments": {"edits": [], "field": "ocr", "item_id": "texte", "base_hash": "texte", "drafted_by": "assistant", "volunteer_code": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• search_concept≈ 1214 tokens

  RETURNS QUOTABLE PASSAGES matched by MEANING (cosine similarity on Gemini embeddings, 768d) — paraphrases and adjacent phrasings match even with zero keyword overlap. PICK THIS when the modern term won't literally appear in historical texts — e.g. "distributed cognition" maps to passages about active intellect, art of memory, wax tablet metaphors; "social contract" maps to pre-Hobbesian discussions of consent and authority. → For exact words/distinctive terms use search_translations (cheaper, more precise); to list which BOOKS cover a topic use search_library; if the user named an author/work, get_book first (semantic search is expensive — reserve it for cross-corpus discovery). Similarity calibration: 0.70+ is a strong match, 0.55–0.70 is worth reading but verify, below 0.55 is mostly conceptual drift. Set max_per_book to diversify results across many books rather than cluster on one source. Each passage carries a snippet_type — quote only "translation" snippets, never "summary". Cross-cultural tip: for pre-modern or non-Western topics, also try source-tradition vocabulary — e.g. for seminal economy try "jing preservation" or "bindu yoga" or "istimnāʾ"; for masturbation try "mollities" (Latin) or "hastamaithuna" (Sanskrit) or "shouyin" (Chinese). The corpus is indexed via period translations that use tradition-internal terminology, so adjacent/euphemistic terms often surface material that modern English keywords miss.

  Paramètres, retour & exemple (14)

  Paramètres

  • lang · string · optionnel
    ISO code of the EDITION to read, e.g. "es". Default "en". Most books have only English — call get_book and read `editions`, or list_books with has_edition, to find the ones that do not. The response always states which edition it served.
  • limit · number · optionnel
    Max passages (default 15, max 50)
  • query · string · requis
    A concept or natural-language description — full sentences are fine (e.g. "tools that extend the mind beyond the body"). Unlike search_translations, this does NOT require words that appear in the corpus.
  • author · array · optionnel
    Keep only passages by these authors, e.g. ["Ficino"] or ["della Porta", "Agrippa"]. Matches by name: every word you give must start a word of the book's author field.
  • year_to · number · optionnel
    Restrict to books published in or before this year.
  • language · string · optionnel
    Filter by a single original language
  • diversity · string · optionnel · tradition | author | off
    How the passages are spread. Default "tradition": each ten holds at most 2 passages per tradition family and per work, so a concept query returns several traditions and not one tradition's nearest pages (a clearly closer match is never passed over). "author": at most one per author and per work, for a cross-author survey. "off": plain nearest-first; use it when looking for one known passage. A quoted query or one naming a year defaults to "off". Each passage carries its book's `tradition` when it has one.
  • languages · array · optionnel
    Filter to any of these languages, e.g. ["Sanskrit", "Arabic", "Chinese"]. Use instead of language when targeting multiple traditions.
  • year_from · number · optionnel
    Restrict to books published in or after this year (filters out modern editions and translations).
  • max_per_book · number · optionnel
    Cap on passages from any single book. Useful when one book dominates the conceptual neighborhood; for one passage per author or per work use max_per_author / max_per_work.
  • max_per_work · number · optionnel
    Hard cap on passages from one WORK (all printings and editions of it count together, via work_id), so seven printings of De occulta philosophia are one witness, not seven.
  • max_per_author · number · optionnel
    Hard cap on passages from one author; the rest are DROPPED (unlike diversity, which only re-orders). Set 1 for "N quotes from N different authors".
  • exclude_authors · array · optionnel
    Drop passages by these authors, e.g. ["Agrippa"]. Use it to iterate a survey: run, see who dominates, exclude, run again.
  • exclude_languages · array · optionnel
    Exclude these languages, e.g. ["Latin", "French", "German", "English"] to surface non-Western sources.

  Exemple d'appel

  {"name": "search_concept", "arguments": {"query": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• search_images≈ 1150 tokens

  Search 200,000+ historical illustrations, emblems, engravings, diagrams, AND 24,000+ artworks (paintings, prints, sculptures). Filter by type, subject, figure, symbol, year. Results interleave two collections: illustrations extracted from book pages (each with a page number and book link) and standalone museum artworks (type: "artwork"). The first few results also return as inline images YOU can see. Hosts that support MCP Apps render an in-chat image gallery for this tool automatically; on other clients images may sit inside the collapsed tool-result view, so never tell the user images are "rendered above" unless the gallery appeared — describe what you see and give each image's url link instead. Every image_url is public and stable — an HTML page that references them directly works in any online browser. If images.length is 0, read the note field — an empty result under a book_id filter means that book has no EXTRACTED images yet, not that the physical book has no plates. A broad query can match tens of thousands (read total): narrow with type/subject/symbol (exact tag strings; look them up with list_image_facets) or page with offset instead of raising limit. On museum-artwork results, a title_is_descriptive flag means the title is an AI description of the picture rather than a title the work was published under — cite such a record by its source_record_title, never by the descriptive one (#4288).

  Paramètres, retour & exemple (11)

  Paramètres

  • type · string · optionnel
    Image type (woodcut, engraving, emblem, diagram). Best-effort: the medium metadata on museum artworks is unnormalized, so treat results as ranked rather than strictly filtered.
  • limit · number · optionnel
    Max results (default 20, max 50)
  • query · string · optionnel
    Text search (e.g., "ouroboros", "tree of life")
  • figure · string · optionnel
    EXACT, case-sensitive match on a figure tag ("Jesus Christ", "putti"). Look it up with list_image_facets.
  • offset · number · optionnel
    Skip this many book-illustration results — page through a large result set instead of raising limit. The response echoes offset and returns next_offset while more remain. Offsets > 0 return the gallery lane only (the museum-artwork lane has no pagination and is included only on the first page).
  • symbol · string · optionnel
    EXACT, case-sensitive match on a symbol tag ("ouroboros", "triangle"). Look it up with list_image_facets.
  • book_id · string · optionnel
    Only return images extracted from this book's pages. Excludes the museum-artwork collection (artworks do not belong to books).
  • subject · string · optionnel
    EXACT, case-sensitive match on an extracted subject tag ("alchemy", "medicinal plants"); a near-miss matches nothing. Look the string up with list_image_facets first, or put loose words in query instead.
  • year_to · number · optionnel
    See year_from — same reliability caution applies.
  • year_from · number · optionnel
    CAUTION: reliable only for book illustrations (source_type "book_illustration"). On museum-artwork records the year field is dirty — some works carry the date a photograph or upload was made (e.g. 2005) rather than the date of the work — so never draw a chronological conclusion from a year-filtered artwork result (#4288 tracks the cleanup). Undated artworks are excluded from any year-filtered search, not passed through.
  • include_thumbnail_base64 · boolean · optionnel
    Embed each result's image as a thumbnail_data_uri (data:image/jpeg;base64,…, ~1000px) directly in the JSON. ONLY useful when your harness consumes tool results programmatically (API/SDK agents that can save the bytes without retyping them) — as a chat assistant you CANNOT copy hundreds of KB of base64 into a file, so do not request this for that purpose. To build a self-contained page from chat instead: fetch the public image_url values with your execution sandbox (if egress is blocked, ask the user to allowlist images.sourcelibrary.org in their network settings), or reference the CDN URLs directly — they are public and stable, so the page works in any online browser. First 6 results only; inline image blocks are suppressed in this mode to keep the payload bounded.

  Exemple d'appel

  {"name": "search_images", "arguments": {"type": "texte", "limit": 1, "query": "texte", "figure": "texte", "offset": 1, "symbol": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• search_library≈ 456 tokens

  RETURNS A LIST OF BOOKS (works on a topic) — NOT passages. PICK THIS to discover which works exist on a subject. → For quotable text use search_translations (exact words) or search_concept (by meaning); if the user already named an author/work, call get_book directly (or list_books to find the ID) — the AI summary + chapter outline is usually the right first answer. Searches titles, authors, subjects, and (as a secondary signal) translated text. Query tips: single distinctive words or short phrases work best ("memory palace", "ouroboros"); quoted phrases match exactly. Each result includes total_matches (full count) + returned (this page) + offset for pagination.

  Paramètres, retour & exemple (8)

  Paramètres

  • sort · string · optionnel · relevance | date_asc | date_desc | title
  • limit · number · optionnel
    Max results per page (default 10, max 100)
  • query · string · requis
    Search query — prefer single distinctive concepts ("alchemy", "tree of life") over long natural-language phrases. Wrap in "double quotes" for exact phrase.
  • offset · number · optionnel
    Pagination offset (use with limit to page through total_matches; default 0)
  • year_to · number · optionnel
    Publication year range end
  • language · string · optionnel
    Filter by original language (e.g., Latin, German, Greek)
  • year_from · number · optionnel
    Publication year range start
  • has_translation · boolean · optionnel
    Only return books with translations

  Exemple d'appel

  {"name": "search_library", "arguments": {"query": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• search_translations≈ 1223 tokens

  RETURNS QUOTABLE PASSAGES (page-level snippets + citation URLs), matched by KEYWORD/term. PICK THIS to find a quote or textual evidence on a topic across the whole library. → If the modern word won't literally appear in historical texts, use search_concept (matches by meaning); to list which BOOKS cover a topic use search_library; to dig inside one known book use search_within_book; if the user named an author/work, get_book first (its AI summary is usually the right first read). Query tips: a distinctive term ("wax tablet") or the sentence you are looking for both work; matching is by word, and a page that prints your words together and in order ranks above one that scatters them. Each passage carries phrase_match (0-1, absolute, not relative to the other results): 1 means the page prints your whole query in order at one place, 0 means your words occur only apart; when every passage is under about 0.2 the library probably does not print that sentence, so try search_concept. It is absent for a one-word query and for a passage found by meaning alone. found_by says which lane found the page: "keyword", "semantic", or "both" (the strongest signal). Each snippet has a snippet_type — "translation"/"ocr" means it is a verbatim extract from the source text; "summary" means it is AI-generated description (do not quote those as the author's words). Response includes total_matches, returned, and offset for pagination. Cross-cultural tip: for pre-modern or non-Western topics, search source-tradition vocabulary rather than modern English terms — e.g. for seminal economy search "jing" or "bindu" or "istimnāʾ", not "semen retention"; for female homoeroticism search "tribade" or "sahq", not "lesbian". The corpus is indexed via period translations that use tradition-internal terminology.

  Paramètres, retour & exemple (15)

  Paramètres

  • lang · string · optionnel
    ISO code of the EDITION to read, e.g. "es". Default "en". Most books have only English — call get_book and read `editions`, or list_books with has_edition, to find the ones that do not. The response always states which edition it served.
  • limit · number · optionnel
    Max results per page (default 20, max 50)
  • query · string · requis
    A distinctive term ("harmony of the spheres", "active intellect") or the sentence you are looking for. A multi-word query matches pages printing ANY of its words, ranked so that pages printing them together come first (see phrase_match); wrap in "double quotes" to require the exact phrase.
  • author · array · optionnel
    Keep only passages by these authors, e.g. ["Ficino"] or ["della Porta", "Agrippa"]. Matches by name: every word you give must start a word of the book's author field.
  • offset · number · optionnel
    Pagination offset (use with limit to page through total_matches; default 0)
  • book_id · string · optionnel
    Search within a specific book
  • year_to · number · optionnel
  • language · string · optionnel
    Filter by a single original language
  • diversity · string · optionnel · tradition | author | off
    Re-order the passages for spread. "author": at most one passage per author and per work in each ten, for a cross-author survey. "tradition": at most 2 per tradition family and per work. Default "off": passages come in relevance order. Nothing is dropped; later pages hold the rest.
  • languages · array · optionnel
    Filter to any of these languages, e.g. ["Sanskrit", "Arabic", "Chinese"]. Use instead of language when targeting multiple traditions.
  • year_from · number · optionnel
  • max_per_work · number · optionnel
    Hard cap on passages from one WORK (all printings and editions of it count together, via work_id), so seven printings of De occulta philosophia are one witness, not seven.
  • max_per_author · number · optionnel
    Hard cap on passages from one author; the rest are DROPPED (unlike diversity, which only re-orders). Set 1 for "N quotes from N different authors".
  • exclude_authors · array · optionnel
    Drop passages by these authors, e.g. ["Agrippa"]. Use it to iterate a survey: run, see who dominates, exclude, run again.
  • exclude_languages · array · optionnel
    Exclude these languages, e.g. ["Latin", "French", "German", "English"] to surface non-Western sources.

  Exemple d'appel

  {"name": "search_translations", "arguments": {"query": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• search_within_book≈ 576 tokens

  SEARCHES INSIDE ONE BOOK (requires book_id). PRIMARILY KEYWORD: it runs a lexical search over the book's pages plus a narrow scoped-semantic pass (top ~10), interleaved by relevance. PICK THIS when you know the wording you are looking for, or want every page of one book mentioning a term. → IF YOU ARE SEARCHING FROM A PARAPHRASE, a half-remembered line, or a modern restatement, USE search_concept INSTEAD — it is the meaning-matching tool and it searches the whole corpus, including translations whose vocabulary differs completely from yours (Thomas Taylor writes "energies" for energeia and "felicity" for eudaimonia, so a sensible modern paraphrase can miss his pages entirely while matching semantically). → To find the book first, use search_library or search_concept, then pass its book_id here. Results are ordered by relevance within this book. Each carries score (0-1, ABSOLUTE, not relative to the other results, so a query the book cannot answer scores low everywhere): the higher of similarity (raw cosine from the meaning pass, calibrated like search_concept: 0.70+ strong, 0.55–0.70 verify, below 0.55 drift) and phrase_match (how much of a multi-word query the page prints in order at one place; 1 is all of it; a one-word query the page prints scores 1). Each also carries found_by ("keyword", "semantic", or "both" — both is the strongest signal). Results flagged is_front_matter are the translator's or publisher's words rather than the author's, and are ordered last. Returns OCR and translation snippets with page numbers, ready to cite.

  Paramètres, retour & exemple (3)

  Paramètres

  • lang · string · optionnel
    ISO code of the EDITION to read, e.g. "es". Default "en". Most books have only English — call get_book and read `editions`, or list_books with has_edition, to find the ones that do not. The response always states which edition it served.
  • query · string · requis
    Search query
  • book_id · string · requis
    The book ID to search within

  Exemple d'appel

  {"name": "search_within_book", "arguments": {"query": "texte", "book_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• share_findings≈ 380 tokens

  Share a research dossier back to the Source Library team — a title, an optional summary, and an ordered list of citations (the passages your thesis rests on). Each citation is a reference { book_id, page, note }, NOT copied text: the library re-renders the canonical quote from the reference, so links stay authoritative. Use this when the user has assembled a thesis backed by passages across one or more books and wants to contribute it back. Like submit_feedback, this goes to the team for review (not an instant public page).

  Paramètres, retour & exemple (5)

  Paramètres

  • name · string · optionnel
  • email · string · optionnel
  • title · string · requis
    Title of the dossier / thesis (2-300 chars)
  • summary · string · optionnel
    Optional prose summarizing the argument (max 5000 chars)
  • citations · array · requis
    Ordered list of supporting passages (1-50). Get book_id + page from get_quote / search_translations.

  Exemple d'appel

  {"name": "share_findings", "arguments": {"title": "texte", "citations": []}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• start_blind_check≈ 258 tokens

  OPTIONAL lane, only for a user who reads the original language and asks to judge pages themselves: up to five random pages from different books, each with scan, transcription and English. Here the USER is the reviewer: ask for THEIR verdict first, before any opinion of yours, then give your reading with exact quotes, let them accept or reject each point, and call submit_page_review with mode "blind" and their final verdict. Never choose the verdict for them. For everyone else use start_review_shift.

  Paramètres, retour & exemple (3)

  Paramètres

  • pages · number · optionnel
    How many pages (1-5, default 5).
  • language · string · optionnel
    Original language the user reads, e.g. "Latin".
  • volunteer_code · string · optionnel

  Exemple d'appel

  {"name": "start_blind_check", "arguments": {"pages": 1, "language": "texte", "volunteer_code": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• start_review_shift≈ 476 tokens

  Start a volunteer REVIEW SHIFT: the user lends their Claude (you) to Source Library to check one book, page by page, against its scans. Use when the user wants to help Source Library, says "start a shift" or "review pages", or asks how they can contribute. The user does NOT need to read the book's language: YOU do the checking (scan against transcription, transcription against English) and the user starts the shift, chooses how many pages, watches and asks questions. Returns ONE page: its scan (inline image), our transcription, our English, the last lines of the previous page, and next (pass to continue_review_shift). Follow first_time once, then how_to_run on every page, working through the stretch of pages without stopping between them. With no book, it resumes the user's last run (pass volunteer_code) or picks a book that needs checking; book takes an id, slug or reader URL if the user names one; new_book true skips the resume. Run shifts only while the user is in this conversation; never set one up to run on a schedule or from a script.

  Paramètres, retour & exemple (4)

  Paramètres

  • book · string · optionnel
    A book the user wants to check: id, slug or sourcelibrary.org reader URL (optional).
  • language · string · optionnel
    Prefer books in this original language, e.g. "Latin" (optional; any language works, the user need not read it).
  • new_book · boolean · optionnel
    True to start a different book instead of resuming the last one.
  • volunteer_code · string · optionnel
    The code from the user's earlier shift, if they have one. Omit on a first shift; one is issued.

  Exemple d'appel

  {"name": "start_review_shift", "arguments": {"book": "texte", "language": "texte", "new_book": true, "volunteer_code": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• submit_feedback≈ 260 tokens

  Submit feedback, bug reports, or feature requests to the Source Library team. Before proposing NEW functionality, read https://sourcelibrary.org/llms.txt and https://sourcelibrary.org/developers — several past submissions proposed building things that already exist (IIIF manifests, Content Search, DTS), which wastes reviewer time. State in the report which docs you checked. Bug reports with record IDs and reproducing queries are the most actionable kind.

  Paramètres, retour & exemple (3)

  Paramètres

  • name · string · optionnel
  • email · string · optionnel
  • message · string · requis
    Your feedback (2-20000 chars). Long structured reports are welcome — the limit was raised from 5,000 because agent reports were pressing against it and being split across submissions.

  Exemple d'appel

  {"name": "submit_feedback", "arguments": {"message": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• submit_page_review≈ 930 tokens

  Record the verdict on one page of a shift. mode "book-run" (start_review_shift / continue_review_shift): the verdict and findings are YOURS, from reading the scan and both texts; set user_looked true if the user checked anything on the page themselves. mode "blind" (start_blind_check): the verdict is the USER'S, given before your reading; put yours in assistant_verdict. Verdicts: both_sound (transcription matches the scan and the English matches the original), translation_drift (transcription right, English departs from it), transcription_off (the text does not match the scan), both_off, unclear (the scan or the text cannot be judged). Quote exact words in every finding. A review is evidence about the page; it does not change it.

  Paramètres, retour & exemple (17)

  Paramètres

  • mode · string · optionnel · book-run | blind
    "book-run" unless the page came from start_blind_check.
  • name · string · optionnel
    How the user wants to be credited (optional).
  • note · string · optionnel
    Anything a later reader should know (optional, max 2000 chars).
  • book_id · string · optionnel
    run.book_id (book run)
  • item_id · string · requis
    page.item_id (book run) or the item's item_id (blind check)
  • verdict · string · requis · both_sound | translation_drift | transcription_off | both_off | unclear
    book-run: your verdict. blind: the user's final verdict.
  • shift_id · string · optionnel
    shift_id (blind check only)
  • page_number · number · optionnel
    page.page_number (book run)
  • user_looked · boolean · optionnel
    book-run: true if the user checked something on this page themselves.
  • text_version · object · optionnel
    page.text_version, passed back unchanged.
  • first_verdict · string · optionnel · both_sound | translation_drift | transcription_off | both_off | unclear
    blind: the user's verdict before your reading, if it changed.
  • volunteer_code · string · requis
  • assistant_model · string · requis
    Your own model name and version, copied exactly as your instructions or system prompt give it, and whether extended thinking is on. Write "unknown" rather than guess; a wrong name is worse than none.
  • reader_languages · string · optionnel
    Languages the user says they read, if they said.
  • assistant_verdict · string · optionnel · both_sound | translation_drift | transcription_off | both_off | unclear
    Your own verdict (blind mode; in book-run it equals verdict).
  • assistant_findings · array · optionnel
    Problems found, each with an exact quote (max 12).
  • verdict_before_assistant · boolean · optionnel
    blind: true if the user committed to a verdict before seeing your reading.

  Exemple d'appel

  {"name": "submit_page_review", "arguments": {"item_id": "texte", "verdict": "both_sound", "volunteer_code": "texte", "assistant_model": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

C'est votre serveur ?

Revendiquez-le sur l'espace créateur (preuve GitHub, domaine ou npm) pour afficher le badge « propriétaire vérifié » et lancer un pentest comportemental à la demande afin d'obtenir le badge « Vérifié ».

Créer un compte / revendiquer →

Source

https://sourcelibrary.org/api/mcp