Changelog

All notable changes to anthroposophy.ai are documented in this file.

The format follows Keep a Changelog. Versioning follows Semantic Versioning.

1.3.0 - 2026-10-21

Reply shape of the subject index, paragraph ranges, per-lecture credit

This release changes the default reply of fetch_index_entry, serves a range of paragraphs, resolves the translator of an English text per lecture rather than per volume, rates English–German paragraph alignment from verified pairings only, labels each lecture's record type (including unknown), and separates the edition's lecture title from a heading the site wrote.

Discipline: a translation_credit with status: not_established names no translator; do not supply one. alignment_confidence is now a statement about verified paragraph pairings; paragraph_aligned: false on a passage means the German beside it is the paragraph of the same number, not a verified counterpart. site_heading is the site's wording, never the edition's; cite a lecture by the edition's title. A record_type of unknown means the edition's note on sources has not been read, not that the text is a stenogram. An index reference with neither a paragraph nor whole was listed before the pinning rule and does not say where in the unit to read.

Added

• fetch_index_entry: detail (compact, headings, full). The default is compact for an entry whose compact reply is at most 20,000 characters and headings otherwise; a headings reply carries the definition section's references. section accepts several ids or headings (a list, or comma-separated in REST). Every reply states its detail and how to ask for the others. REST twins: GET /v1/steiner/index and GET /v1/steiner/index/<entry>?detail=&section=.
• fetch_passage: through (a later paragraph of the same unit, at most 12 paragraphs); the reply names the span as GA n/<unit>/pNNN-pMMM with extent: "range" and carries the aligned other-language paragraphs as aligned_witness where pairings exist. Span URLs /v1/<corpus>/<lang>/ga/<ga>/<unit>/pNNN-pMMM.json and the HTML form; passage_json passes window and through. Lecture pages carry a three-digit anchor per paragraph (#p003); old #p3 links are redirected on the page.
• translation_credit.status (established, not_established, with reason) resolved per unit from the lecture's own source page; witness_quality values not_established, automated, original_english.
• unit_kind on passages and lecture listings of suffixed units (second_lecture_same_day, second_translation with same_lecture_as and german_repeats, duplicate_text, second_piece_same_name, same_lecture_second_id, undetermined). Unit ids are unchanged.
• site_heading (nullable) and site_heading_note on every reply that names a unit; /about/site-headings/ explains the field. Lecture pages show the edition's title with the site heading as a marked subtitle.
• record_type: "unknown" on every lecture unit with no record type on file; record on fetch_lecture; evidence_source on volume rows and source on unit rows of the record-type data. Subject index rows show a record unknown badge, the place of delivery and the translator (or "no translator on record").
• alignment_reason values from the pairing loader (beads:<method>, identity_contradicted, paragraph_counts_match); fetch_parallel states each edition's reason.
• Index references carry extent (paragraph, range or unit), through on a range (at most 12 paragraphs; its links go to the span) and whole: true where the unit treats the heading's subject throughout; fetch_index_entry compact rows gain a sixth column, whole; the record schema (/index/data/schema.json) lists the three fields.
• section_aliases on index entries: names that lead to one heading of an entry rather than the whole entry (lesser guardian, greater guardian and their German forms, in guardian-of-the-threshold). fetch_index_entry given such a name returns that section with leads_to_section and leads_to_note; the index_entry pointer on search_steiner carries the section.
• A unit served under a slug can also be resolved by its date when exactly one unit carries that date and none is served under it (GA 98, 7 November 1907).

Changed

• fetch_index_entry default reply is compact (was the full record); compact and headings replies are minified JSON. Entry lookup folds ae/oe/ue/ss and umlauts both ways, ignores case and a leading article.
• alignment_confidence is taken from verified paragraph pairings; a rating that rests only on equal paragraph counts is at most medium. paragraph_aligned is true only where a pairing exists.
• translation_credit is resolved per unit; a volume's credit is no longer applied to its lectures.
• Translator's endnotes are served as notes with speaker the translator.
• Server instructions and tool descriptions updated for the above; mirrored to /llms.txt. MCP server version and SERVER_VERSION are 1.3.0.

Fixed

• Website navigation text at the tail of the GA 1f German files was served as passages; removed.
• GA 254, 10 October 1915 (German): two paragraphs joined by a column mark are split; later paragraph numbers in that unit move by one.
• GA 254 (German): misreadings corrected per the repair manifest.
• GA 82, 12 April 1922 (German and English): a paragraph repeated in the source text is served once; later paragraph numbers in that unit move down by one.
• GA 162, 24 May 1915 (English): paragraph 33 held the lecture's text twice and now holds its own text only; a misreading (& for a letter) is corrected in both languages.
• Sentences split across two paragraphs by a page break or a picture line are joined (GA 115, 4 November 1910; GA 35; further units found by the same scan), and paragraphs that repeated another paragraph's text are removed (GA 28, 29, 70b, 326); later paragraph numbers in an affected unit move down. Subject-index references follow the new numbers.

1.2.1 - 2026-10-19

Register of the Gesamtausgabe

This release makes the register of the Gesamtausgabe at /contents/ available to agents. The register lists every item of the edition that the publisher's catalogs, the archive's completion sheet, AnthroWiki or the English bookseller name, whether or not the site serves its text: the numbered volumes, numbers published only as sub-volumes, empty and superseded numbers, the art volumes of the K series, the digital notebook edition, announced companion volumes and the Beiträge booklets. Each record states whether the site serves German and English, the source of each served text, the editions and translations known, and the witness for every fact.

• Number lookup. fetch_ga_register resolves a GA number in any written form (GA 266/I, 266a, K 58/1, eGA 47/48, Heft 24) to its register record.
• Absence explained. A search_steiner call scoped to a volume the corpus does not hold carries register_entry, stating what the number is: an empty number, an art volume, a volume not served.

Discipline: the register is bibliographic data, never Steiner's words; fetch passages before saying what a volume contains. Translations listed as known but not served are bibliography, not text the server holds. Where witnesses disagree, the record gives the candidates side by side.

Added

• fetch_ga_register MCP tool: ga (a number in any written form; * lists the edition, beitraege the booklet series) and optional scope (audited, parent, empty_number, superseded_number, art, digital_edition, companion, beitraege). Every fact is {value, w}, w naming keys of the record's witnesses. A miss returns not_in_register with did_you_mean.
• register_entry on search_steiner responses with absence_signal='volume_not_held': the number's scope, titles, served state, and its HTML and JSON URLs.
• Register data at /contents/data/: ga.json (the listing), one <slug>.json per item (the number with / written -), beitraege.json and ga.csv.

Changed

• Server instructions: bibliographic questions about a GA number are answered from fetch_ga_register. Mirrored to /llms.txt.
• MCP server version and SERVER_VERSION are 1.2.1.

1.2.0 - 2026-10-09

Concept index

This release adds a concept index at /index/. Each entry lists the chapters and lectures of the Gesamtausgabe where its concept is treated, under sub-headings: Steiner's own books (GA 1 to 28) first, then articles (GA 29 to 46), then lectures in date order. An entry contains no quotations. Every reference is a locus verified against the served corpus when the entry is built; a paragraph number is given only where one passage is the point.

• Orientation before keyword search. Unscoped searches whose query names an indexed concept, by its headword or an alias in English or German, carry an index_entry pointer ahead of the hits. Search ranks by word frequency; the entry lists where the concept is treated.
• Entry retrieval. fetch_index_entry returns an entry by id, headword or alias, whole or one sub-heading at a time. Each reference carries the GA volume and its title, the served unit id, language, edition, witnesses, record type where the publisher states one, and the viewer and JSON URLs for fetching the text.

Discipline: an entry says where to read, not what the text says. Its headings and scope lines are the index's wording, never Steiner's; quote only what fetch_passage returns. entry_not_found means no entry exists yet, not that the corpus is silent on the concept.

Added

• /index/: one page per entry, served as HTML and as JSON at /index/data/<id>.json, with the record schema at /index/data/schema.json (index-entry/1.0) and a listing at /index/data/index.json. The HTML page shows the fields of the JSON record and nothing else.
• fetch_index_entry MCP tool: entry (id, headword or alias; * lists the entries) and optional section. A miss returns entry_not_found with did_you_mean.
• index_entry on search_steiner, search_anthroposophy and /v1/steiner/search responses when the query names an indexed concept and no ga, ga_range or date filter is set: the entry's id, headwords, sub-headings with reference counts, and its HTML and JSON URLs. Placed first in the envelope.
• index_entry on a trace_claims node_not_found response when the node names an indexed concept.

Changed

• Server instructions: concept-level questions start from fetch_index_entry, and the index_entry pointer is read before the hits. Mirrored to /llms.txt.
• Site navigation includes Index.
• MCP server version and SERVER_VERSION are 1.2.0.

1.1.5 - 2026-08-27

Research workflows

This release makes the server's negative answers say what they are answers to, and answers three questions quote_safe never answered: who is speaking in a passage, how the volume was recorded, and how good a witness a translation is. The work reached production in three same-day deploys, briefly numbered 1.1.5 to 1.1.7; it is one release.

• Search for two words near each other. word NEAR/10 word now executes. It was previously rejected as invalid syntax, rewritten into literal words, and the zero returned as an authoritative co-occurrence absence.
• Tell a rewritten query from the one you sent. Responses carry query_as_run. A sanitizer-rewritten query returns absence_signal='query_rewritten', which is not authoritative for the original; relaxation still runs on the content words.
• Read what a phrase zero settles. A phrase or proximity zero keeps its verdict and empty hits and adds a relaxed block: operators dropped, content words swept, and term_document_counts per word — which separates 'never occur together' from 'absent'.
• See how much of the original a translation holds. witness_completeness gives the witness's word ratio to the original, banded, on passages, lectures and volumes. A translation that omits passages while reading as continuous prose is detectable before it is quoted from.
• Tell who is speaking. authorship resolves the passage: a labelled transcript turn, prose written about Steiner in the third person, an editor's footnote, or a character's line in a drama — two of those volumes being not his composition at all.
• Weigh the record and the translator. record gives the volume's record type where the publisher's subtitle states it. witness_quality names the translator or reports none recorded, and compares the passage's load-bearing names with its aligned German.
• Read a German snippet. Snippets were rendered from the normalized index, so German arrived stemmed — der zweit brief wird entsiegelt. They are now cut from the source text with the matched words marked.
• Recognise a second locus as the same witness. 195 groups of lectures hold byte-identical text under two or three ids. Hits and passages carry also_served_as.

Discipline: a 0-hit is authoritative that the queried wording cannot be quoted, never that the corpus lacks the idea; the instructions now say so where they previously said the opposite. quote_safe certifies the bytes only — speaker, record type, completeness and translation fidelity are separate fields. A clean entity_parity checks names alone, and a mismatch can equally mean a wrong alignment. Snippets remain non-quotable, and two loci in also_served_as are one witness, not two.

Added

• query_as_run on every search response, and three absence_signal values: query_rewritten (the sanitizer rewrote the query, so the zero is not authoritative for the original), volume_not_held (the ga= filter names a volume not in the corpus), translation_may_be_abridged (a ga=-scoped zero in a translation under 75% of the original).
• relaxed block on phrase, proximity and boolean zeros: operators_dropped, term_document_counts, and broad with a note when a single-word sweep exceeds 500 passages. Relaxed passages are never merged into hits.
• witness_completeness on fetch_passage, fetch_lecture and the volume endpoint: ratio, band, may_be_abridged and the two word counts. Bands are severe below 0.50, short below 0.75, watch below 0.85, comparable to 1.35, fuller_than_original above; under a 500-word original the ratio reads too_short_to_judge. verify_locus warns when a witness of the locus is abridged.
• authorship resolves the speaker at passage level from four signals, most specific first: an editor's [^n]: footnote, a transcript speaker label, the per-lecture map, third-person prose. New kind values dialogue, report (is_steiner: null), dramatis_persona (with author), adapted_work (GA 42, GA 43) and editorial. Labels come from a map of those recurring three or more times in their volume.
• record on hits and fetch_passage: volume record type — stenogram, listeners_notes, own_manuscript, mixed_sources — with the publisher's subtitle as evidence. An absent record means unknown.
• witness_quality on translation passages: translator (or status='unknown', which includes a credit naming only a publisher) and entity_parity against the aligned German, with missing_from_translation, bead_confidence and a likely_cause. Passages under 25 words, or with no alignment bead, read not_checkable.
• also_served_as: the other lecture ids carrying byte-identical text, noted as the same witness rather than independent attestations.

Changed

• Server instructions: a 0-hit is described as authoritative that the wording cannot be quoted, not that the phrasing is absent from the corpus, and the authorship.kind vocabulary is documented. The byte-for-byte quoting rule is unchanged. Mirrored to /llms.txt.
• Phrase, proximity and boolean zeros are now relaxed into the relaxed block; such queries were previously never relaxed. hits, absence_signal and the verdict are unchanged. German phrase zeros sweep the query's capitalized nouns first.
• Search snippets are rendered from the passage text rather than the normalized index, anchored on terms of four characters or more so articles do not draw the marks. snippet_is_verbatim remains false.
• Advice arrays are tiered per MCP session: quotation and attribution lines accompany every response, the remaining guidance is sent on each tool's first call per session.
• german_anchor on an unaligned bead reports NOT ALIGNED and notes the German may hold the content under different numbering; it previously reported a gap in the German original.
• MCP server version and SERVER_VERSION are 1.1.5. The intermediate 1.1.6 and 1.1.7 values, served for about an hour during the day's staged deploys, describe this same surface.

Fixed

• word NEAR/N word raised a syntax error, was rewritten by the sanitizer into literal words including the operator itself, returned zero, and the zero was certified as verbatim co-occurrence absence. The infix form is rewritten to NEAR(a b, N) and executed; the tool documentation, which called NEAR both unsupported and supported, is consistent.
• A ga= filter naming a volume absent from the corpus returned a missing-language message listing no available languages.
• A society annual report, a questioner's turn, an editor's footnote and a character's line in a drama were served with no authorship field, which by the corpus's stated default meant attributed to Rudolf Steiner.
• An edition credit naming a publisher with a null translator was served as a filled-in credit.
• German snippets were unreadable: the index stores stemmed, case-folded, diacritic-stripped text and snippet() rendered from it.
• fetch_passage raised an operational error against a database without the paragraph_align table.

1.1.4 - 2026-07-06

Research workflows

This release serves quote-level verification — the byte-for-byte quotation discipline as an API operation rather than a client-side re-implementation — and fixes the bare-/mcp connector failure.

• Verify a quotation before printing it. verify_quote on a candidate inherited from the secondary literature returns near_miss with a word-level diff against the closest corpus window and the verbatim span at its locus — an inherited misquote is caught in one call instead of propagated.
• Audit a document’s quote ledger in one call. Batch mode takes up to 50 quote+locus pairs and returns per-item verdicts plus an all_quote_safe summary — a published artifact’s “verified” claim can be the API’s output rather than the author’s promise.
• Locate a quote whose wording differs across editions. Without a locus, verify_quote locates candidates corpus-wide and checks every (lang, edition) witness, so a quotation taken from any indexed translation resolves to its locus.
• Review an elision’s surroundings in one call. fetch_passage with window=N returns the passage plus up to N neighbors per side under context.before / context.after — checking that an ellipsis in a spliced quotation hides nothing meaning-inverting no longer costs one call per neighbor.

Discipline: only quote_safe: true clears a candidate for quotation marks. match_normalized means the wording matches but the typography does not — quote match.verbatim_text exactly as returned, never your candidate. near_miss means the corpus reads differently; quoting the candidate propagates drift. A locus-scoped no_match is authoritative for that locus; a corpus-search no_match is strong evidence of absence, not proof.

Added

• Tool verify_quote: quote-level verification. verify_locus proves a locus exists; this proves a quotation matches. Single mode takes candidate with optional scope (ga + lecture (+ passage_number), lang, edition); batch mode takes quotes[] (max 50) and returns per-item results plus a summary with all_quote_safe. Verdicts, strongest first: match (byte-for-byte substring of a witness; quote-safe), match_normalized (matches after typographic normalization only — quote style, dash family, whitespace runs, emphasis markers; not quote-safe; match.verbatim_text carries the exact span to quote instead), match_spliced (candidate contains ellipses; every fragment matched in order within one lecture; quote-safe only when every fragment matched byte-for-byte), near_miss (closest corpus window, with similarity and a word-level differences diff; near_miss.corpus_text carries the verbatim span), no_match. Without a locus the check locates candidates corpus-wide via full-text search (exact phrase first, then bag-of-words with 0-hit relaxation); a locus-scoped no_match is authoritative for that locus, a corpus-search no_match is strong evidence of absence, not proof. All (lang, edition) witnesses are checked by default, so a quotation from any indexed translation can be located. REST equivalent: POST /v1/steiner/verify-quote. Deployed 2026-07-06.
• fetch_passage parameter window (integer 0-5, default 0): returns up to N neighboring passages per side — same lecture, same lang + edition — under context.before / context.after, each with verbatim text, locus, and URL. One call to review the surroundings of a passage, e.g. the text elided by an ellipsis in a spliced quotation. Also on the REST passage endpoints as ?window=N. Deployed 2026-07-06.

Changed

• MCP tools/list returns 19 tools (5 GA + 7 FMC + 5 sources + 2 claim-registry).
• Server instructions extended: candidate quotations — especially those inherited from secondary literature or model memory — should be checked with verify_quote before being placed inside quotation marks; only quote_safe: true clears a candidate.
• MCP server version and SERVER_VERSION bumped to 1.1.4.
• Per-IP rate limiting is now enforced on both the REST and MCP surfaces (app-level token bucket, ~240 requests/minute/IP). Over-limit requests receive 429 Too Many Requests with a Retry-After header. This is generous flood protection, not a precise quota — a considerate client rarely encounters it; batch where the surface supports it (verify_quote ledger mode) and prefer targeted fetches over sweeps.
• Search boolean operators are now normalized operator-aware: query operands are stemmed and case-folded while uppercase OR/AND/NOT/NEAR, quoted phrases, prefix *, and column filters are preserved. Previously the query normalizer lowercased the whole query, demoting FTS5’s uppercase-only operators to literal terms, so an OR union became an implicit-AND and could return an authoritative zero where the individual terms matched — a manufactured verbatim-absence. A 0-hit now reports absence_signal all_alternatives_absent for an OR union (every alternative absent) distinctly from exact_phrase_absent for a phrase/AND/NEAR co-occurrence miss. Related surface additions: two_voice_volume on search hits and fetch_passage (with an attribution_warning on the passage) flags correspondence volumes where a passage may be a letter by a correspondent rather than by Steiner; edition_coverage ({lang: [editions]}) accompanies any single-ga search and fetch_passage, and a 0-hit in a language a volume lacks returns absence_signal=language_not_available rather than topical absence; fields=compact (REST ?fields=compact) drops per-hit available_versions[] and edition metadata for large result sets; and ga with from_date/to_date is documented as the scoping recipe for locating a passage within a long lecture. Deployed 2026-07-07.
• Search hits and fetch_passage now carry an authorship object on the corpus’s non-Steiner-bearing loci — correspondence letters, question-and-answer / discussion sessions, and editorial apparatus — resolving the speaker at lecture granularity, finer than the volume-level two_voice_volume flag. authorship.is_steiner is false with a named speaker where a passage is authored by a correspondent, a questioner session, or an editor (not to be quoted as Steiner), true for Steiner’s own hand (the attribution_warning is then suppressed), and null where the source leaves the hand open; an absent authorship field means the passage is attributed to Steiner. This narrows attribution from the whole volume to the individual lecture, so a question-and-answer session or a correspondent’s letter is flagged even in a volume the coarse flag does not cover. Deployed 2026-07-07.

Fixed

• MCP transport: requests to the bare /mcp path (no trailing slash) received 307 Temporary Redirect to /mcp/. MCP clients that do not follow redirects never completed initialize, so connectors configured with the bare URL attached no tools. Both /mcp and /mcp/ now serve the streamable-HTTP transport directly, with no redirect hop. The redirect had been present since the MCP endpoint first shipped; it surfaced when connector-side redirect handling changed. Deployed 2026-07-06.

1.1.3 - 2026-06-11

Research workflows

This release adds schema-set enumeration and documents the claim-registry tools deployed 2026-06-04.

• Enumerate the FMC schema set. list_fmc_schemas pages the full set in deterministic id order (total_count plus offset/limit echo). Schemas were previously retrievable only individually by FMC00.NNN id; no enumeration path existed.
• Filter schemas by topic or id range. query matches description text and linked-page slugs (descriptions are sparse upstream; linked pages carry most of the topical signal). prefix scopes an id range, e.g. prefix="FMC00.5".
• Image retrieval handoff. Entries with image_only: true carry no usable description; pass the same schema_id to fetch_fmc_schema_image for the diagram as inline ImageContent. fetch_fmc_schema remains the single-schema detail record.
• Trace registered claims. trace_claims(node) returns registered claims touching a concept or work: typed relations with up to three verbatim anchors each (leg_a, leg_b, witness), every anchor carrying its locus, edition, content hash, and fetch URL. mode="divergence" restricts results to divergence claims.
• Re-verify a claim before asserting it. verify_claim(claim_id) re-runs a registered claim against the serving corpora: each anchored leg is checked by content-hash (registry integrity) and re-fetch (corpus drift). Per-leg verdicts: anchor_present, anchor_missing, hash_mismatch, fetch_error.

Discipline: list_fmc_schemas entries are curatorial_content — cite as ‘FMC schema FMC00.NNN (FMC curator, anthroposophy.eu, CC-BY-SA)’; descriptions are paraphrase and never go inside quotation marks. For claims, quote only anchor_text, byte-for-byte, and report relation types as given. node_not_found from trace_claims means no claims are registered for that node, not that the corpora are silent — use search_steiner / search_sources for frontier research. The registry’s selection principle (steinerian-anthroposophical) is named in every envelope.

Added

• Tool list_fmc_schemas: enumeration and filtered discovery over the FMC schema set. Filters: prefix (schema-id prefix), query (case-insensitive substring over description text and linked-page slugs). Per-schema fields: schema_id, description (null when none exists), schema_image_url, linked_pages, page_url, and image_only (the signal to call fetch_fmc_schema_image). Deterministic schema_id ordering with total_count and offset/limit echo. Closes the schema-discovery gap: the wiki’s schema-listing page is outside the mirror’s indexed namespace, so fetch_fmc_page could not reach it.
• Tool trace_claims (deployed 2026-06-04; documented here): registered claims touching a concept or work — typed relations across the corpora (Steiner GA, Bible, Aquinas, Church Fathers, Bhagavad Gita, and the other /sources/ works). Each claim carries a relation type (direct_citation, commentary_on, doctrinal_appropriation, thematic_affinity, ambient_influence, doctrinal_contrast, interpretive_continuation, complementary_polarity), a mode_tag (resonance | divergence), a statement, and up to three verbatim anchors (leg_a, leg_b, witness) with loci, editions, content hashes, and fetch URLs. relation_source identifies the asserting layer (curated / registered / citation_index / summary_engagement); machine-derived edges are weaker evidence than curated ones. The registry is compiled from reviewed claim files and deployed as an artifact; no write path is exposed on the public surface.
• Tool verify_claim (deployed 2026-06-04; documented here): live re-verification of a registered claim. Each anchored leg is checked by content-hash against the registry (integrity) and re-fetch against the serving corpora (drift). Per-leg verdicts: anchor_present, anchor_missing (treat the claim as unverified), hash_mismatch (registry integrity failure), fetch_error (infrastructure, not drift). ok=true means every anchored leg verified at verified_at. Claim ids come from trace_claims envelopes (edges[].claim_id).

Changed

• MCP tools/list returns 18 tools (4 GA + 7 FMC + 5 sources + 2 claim-registry). Corrects the 1.1.2 entry, which stated 14; the count at 1.1.2 was 15 (search_anthroposophy omitted from its FMC arithmetic).
• MCP server version and SERVER_VERSION bumped to 1.1.3. The server version field now references the SERVER_VERSION constant instead of carrying a second literal; the two values had drifted apart twice before 1.1.2.

1.1.2 - 2026-05-23

Research workflows

This release adds the /sources/ corpus. It is curated by Steiner-engagement, not by canon — these are the foundational texts the GA itself takes up.

• From a Steiner lecture to its source text. find_source_chapters_for_ga_locus("GA 142, 1912-12-28") returns the Sources chapters the curator has linked to that lecture (Bhagavad Gita Discourses, Mahabharata Parvas, the Pauline epistles as the GA 142 counterpart text). fetch_source_chapter then quotes the verses verbatim.
• From a foundational text to Steiner’s treatment of it. fetch_source_summary on the Kaṭha Upanishad or De Anima Book II names the verified GA loci where Steiner engages it; fetch_passage returns his actual words.
• Cross-corpus surveys in a single call. search_anthroposophy(query=..., corpus="all") fans GA + Free Man Creator (FMC) + Sources together, distinguishing hits by [ga] / [fmc] / [sources] tags with per-corpus totals.
• Honest gap-detection. When the curator has examined a chapter and Steiner did not engage it, steiner_status: "not_engaged" marks the absence explicitly. Aristotle’s De Anima carries this flag.
• Stream-filtered comparative work. search_sources(stream="indian") scopes to Vedic / Upanishadic / Gita / Mahabharata material; stream="greco-christian" to Patrologia + Aristotle + Aquinas + Hermetic Corpus; stream="western-european" to the Beguine + Rhineland + Theosophical + Goethean strands.
• Curator’s cross-tradition congruences as named syntheses. Each summary surfaces the FMC curator’s explicit cross-tradition observations (e.g. De Anima’s tripartite soul ↔ Vedanta’s pancha-kosha ↔ Plotinus’s nous-triad ↔ Kabbalah’s nefesh / ruach / neshamah) as paraphrase the curator owns, never as quotes from those traditions.

Discipline: only fetch_source_chapter is quote-safe (quote_safe=true). fetch_source_summary, find_source_chapters_for_ga_locus, and every steiner_loci.statement field are FMC-curatorial paraphrase (quote_safe=false) — they capture the verified loci but they never go inside quotation marks.

Added

• /sources/ corpus integration — Sacred, Inspired and Foundational Texts across 5 wisdom-streams (Indian, Persian, Egyptian-Hebrew, Greco-Christian, Western-European). Vedic scripture, the Greek-philosophical tradition (Aristotle, Plato, Plotinus), Patrologia (Church Fathers), Aquinas (Summa Theologiae + Summa Contra Gentiles), Beguine + Rhineland mysticism, Hermetic Corpus, Rosicrucian Manifestos, Theosophical inheritance (Isis Unveiled, Secret Doctrine, Mahatma Letters), Goethean works.
• Tool search_sources: full-text search over /sources/ chapters. Filters: work, lang, edition, stream, tradition, form, year_from, year_to. Returns ranked chapter hits with chapter_locus, snippet, curatorial blurb, and summary_url (when present) pointing at the FMC-curatorial paraphrase block.
• Tool fetch_source_chapter: retrieve verbatim chapter text by chapter_locus (<work>[/<subwork>...]/<chapter>). Returns text_trust='verbatim', quote_safe=true, parent breadcrumbs, descriptor subtitle + blurb, and available_versions[]. Use to obtain quote-safe text from /sources/.
• Tool fetch_source_summary: retrieve the FMC-curatorial paraphrase block for a chapter or work — theme, soul-faculty, Steiner engagement (paraphrase + verified GA loci, or the honesty-gap sentinel “not engaged in the GA corpus”), cross-tradition congruences. text_trust='paraphrase', quote_safe=false. The structural verbatim-vs-paraphrase split that prevents accidental quotation of curator-summary content.
• Tool list_source_works: discovery — enumerate /sources/ works by stream, tradition, form, parent_locus, top_level_only, has_summary. Returns work_locus, name, tradition, year_approx, chapter_count, word_count, landing_url, summary_url.
• Tool find_source_chapters_for_ga_locus: reverse cross-reference (sources → GA). Given a GA locus (GA 142, 1912-12-28) or volume (142), returns which /sources/ chapters cite that lecture, with paraphrase statements. Driven by the sources_ga_xref bridge, populated from chapter and work summaries.
• search_anthroposophy umbrella extended with corpus='sources' and corpus='all'. 'all' fans GA + FMC + sources in a single call with round-robin interleave; 'both' remains the back-compat default (GA + FMC). Response carries sources_total alongside ga_total and fmc_total.
• Tool fetch_fmc_schema_image: returns inline ImageContent for FMC schemas, mirror-first with upstream fallback. Resolves the image-fetch gap where MCP clients could not follow MediaWiki redirects to the upstream JPEGs.

Changed

• Top-of-document Quotation discipline (read first) block in server instructions extended with a /sources/ paragraph. fetch_source_chapter returns quote-safe text (text_trust='verbatim'); fetch_source_summary returns FMC-curatorial paraphrase (text_trust='paraphrase', quote_safe=false) — never quote, even when paraphrasing is tight. Each steiner_loci entry inside a summary names a verified GA locus retrievable via fetch_passage; the surrounding statement is the curator’s paraphrase, not Steiner’s words. find_source_chapters_for_ga_locus returns paraphrase as well.
• search_anthroposophy.corpus enum: ['ga','fmc','both'] → ['ga','fmc','sources','both','all']. Tool description acknowledges sources as a third corpus alongside GA and FMC.
• MCP tools/list returns 14 tools (4 GA + 5 FMC + 5 sources).
• MCP server version and SERVER_VERSION bumped to 1.1.2. Both were stale prior to this release (1.1.0 and 1.0.0 respectively).

Fixed

• scripts/deploy_site.sh parity probe: added retry/backoff (8 attempts, 0.5s × attempt#) to tolerate uvicorn’s multi-worker spawn window. systemctl is-active returned active before all workers had bound to port 8044, causing a false rollback on the 2026-05-22 deploy.
• v2/server/main.py: removed a stale import sources_shared that crashed all 4 anthro-api workers with ModuleNotFoundError on startup (the module is on the MCP service’s path, not the API service’s). The /sources/ REST endpoints remain deferred per the MCP-first pattern; the import was dead code.
• find_fmc_pages_for_ga_locus.inputSchema: removed the top-level anyOf that required either ga_locus or ga_volume. Anthropic’s custom tool API rejects top-level anyOf / oneOf / allOf in input_schema; any agent call that included this tool in its toolset was failing with 400 tools.N.custom.input_schema: input_schema does not support oneOf, allOf, or anyOf at the top level. The field-presence constraint is now enforced in the dispatch handler instead (which already validated the same rule and returned error_code: "missing_argument"). A regression test (test_no_top_level_schema_combinators) sweeps every FMC tool schema for top-level combinators.

1.1.1 - 2026-05-18

Research workflows

This release adds the quotation-discipline layer for corpus citations.

• Anything inside quotation marks must be byte-for-byte from a corpus retrieval. fetch_passage, fetch_source_chapter, or a quoted_span record with status=stored — minor trimming, smoothing, or dropping a single word counts as paraphrase, even when self-labelled as quoting.
• 0-hit responses carry an authoritative “do not quote” signal. search_steiner / search_anthroposophy / search_sources return corpus_verdict: "NO_MATCH — ..." in the response payload when total == 0. The verdict is authoritative: the queried text is not in the indexed corpus.
• Bare GA-volume-and-year citations are rejected as unverified. A citation such as “GA 26, 1924” without a specific lecture-date or passage-number is not a verified citation.
• Reverse-lookup questions get the dedicated tool. “Which FMC pages cite GA 146, 1913-06-02?” routes to find_fmc_pages_for_ga_locus, not generic FTS — exact-locus questions answered by an exact-locus tool.

Citations in agent output therefore reflect retrieval rather than training memory; a “not in corpus” answer is authoritative.

Added

• corpus_verdict field on search_steiner and search_anthroposophy responses when total == 0. Carries an authoritative “do not quote” signal in the response payload, so the agent reads it directly rather than relying only on session-level instructions.

Changed

• New top-of-document Quotation discipline (read first) block in server instructions. Requires search_steiner → fetch_passage before opening quotation marks. 0-hit corpus results are authoritative. Bare GA-number-plus-year citations (e.g. “GA 26, 1924”) rejected as unverified. Byte-for-byte means byte-for-byte: dropping a single word (‘thus’, ‘then’, ‘also’) or trimming for slide-readability breaks verbatim status; self-labeling a paraphrase doesn’t exempt it.
• Additional agent-discipline blocks in server instructions: inline citation rendering at parity with web-source linking; section references must use <page_slug>#<section_anchor> form; reverse cross-reference questions must use find_fmc_pages_for_ga_locus; stay inside the corpus when verify_locus returns true; citation follow-ups must verify both page and section anchor before answering.
• Tool find_fmc_pages_for_ga_locus: description lead rewritten around the question pattern it handles (“which FMC pages cite this lecture?”); routes away from search_anthroposophy for reverse-lookup queries.
• Tool search_anthroposophy: closing “When NOT to use this tool” block redirects exact-locus reverse lookups to find_fmc_pages_for_ga_locus.
• Shared FMC source-discipline contract (in search_anthroposophy + fetch_fmc_page descriptions): closing Quotation-marks discipline sentence reserves quotation marks for quoted_span records with status='stored' or verbatim Steiner passages from fetch_passage.

Fixed

• find_fmc_pages_for_ga_locus.inputSchema now declares anyOf requiring ga_locus or ga_volume. Was permissive on {}.
• Removed dead corpus argument from search_steiner / fetch_passage / fetch_parallel dispatch. Handlers pass corpus="steiner" as a literal. fetch_parallel.inputSchema no longer advertises the single-value corpus enum.

1.1.0 - 2026-05-14

Research workflows

This release adds the Free Man Creator (FMC) curatorial layer (anthroposophy.eu — pages, schemas, structured quoted-span records, and GA cross-references) to the MCP surface alongside the GA.

• Reverse cross-reference (GA → FMC). find_fmc_pages_for_ga_locus("1913-06-02-GA146") returns the FMC pages citing a GA locus, with attribution class (adjacent / section-inferred / main-body-reference).
• Curatorial page retrieval. fetch_fmc_page("Krishna_impulse") returns ordered curatorial chunks (lead, sections, subsections, discussion notes) plus quoted_span records with provenance.
• Schema retrieval. fetch_fmc_schema("FMC00.389A") returns description, image URL, linked pages, and cited GA loci.
• Link-graph traversal. traverse_anthroposophy_links runs BFS over the FMC cross-link graph (depth 1-3, configurable max-nodes).

FMC content is CC-BY-SA paraphrase; quoted_span records carry the curator’s structured provenance (status=stored for inline verbatim, status=referenced for GA pointers that resolve via fetch_passage). The unified search_anthroposophy umbrella ranks GA + FMC hits together for queries spanning both.

Added

• FMC (anthroposophy.eu) corpus integration. The curatorial wiki at anthroposophy.eu is now retrievable through the MCP connector alongside the Gesamtausgabe. License: CC-BY-SA.

• Tool search_anthroposophy: unified full-text search across the GA and FMC corpora. Filters: corpus, chunk_class, target_corpus, status, page_slug. Inherits GA filters (lang, edition, ga, ga_range, from_date, to_date). Detects ranking mode from query content (primary_seeking, comparative, curatorial, default) and adjusts interleave order accordingly.

• Tool fetch_fmc_page: retrieve an FMC page by slug. Returns ordered curatorial chunks (lead, sections, subsections, discussion notes), quoted_span records with provenance, and the page’s GA cross-references.

• Tool fetch_fmc_schema: retrieve a schema by FMC00.NNN id. Returns description, image URL, linked pages, and cited GA loci.

• Tool traverse_anthroposophy_links: BFS over the FMC link graph from a root page slug. Depth clamped to [1, 3]; configurable max_nodes.

• Tool find_fmc_pages_for_ga_locus: reverse cross-reference. Given a GA locus (1913-06-02-GA146) or volume (146), return all FMC pages that cite it with attribution class (adjacent, section-inferred, main-body-reference).

• Source-discipline contract enforced in tool response structure. quoted_span records with status=referenced carry no verbatim text; agents follow target_locus to the primary corpus. Records with status=stored carry the full embedded_attribution chain. All FMC retrieval results carry primary_anchor (corpus, page_url, curator, license) and content_hash for citation stability.

Changed

• MCP tools/list returns 9 tools (4 GA + 5 FMC).

1.0.0 - 2026-05-07

Research workflows

Initial release. The complete Gesamtausgabe (GA 1 through GA 354, public domain) behind a single MCP surface.

• Search the full corpus by keyword. search_steiner runs FTS5 with Porter stemmer + diacritic normalization. Filters: lang, edition, ga, ga_range, from_date, to_date.
• Verify a GA reference exists before citing. verify_locus confirms (ga, lecture_date) is in the corpus — useful as a citation gate before publishing or quoting.
• Retrieve any passage verbatim. fetch_passage(ga, lecture, passage_number, lang, edition) returns text + stable citation URL in either German original or available English translation.
• View parallel translations side-by-side. fetch_parallel returns multiple editions of the same locus; per-edition translator + year + basis are surfaced, and an alignment_confidence flag warns when paragraph counts diverge between source and translation.

German originals and English translations are cross-retrievable per locus. Browseable HTML at /v1/steiner/<lang>/ga/<vol>/<date>/p<NNN>, with content-negotiated JSON via the .json suffix. The MCP surface is anonymous and per-IP rate-limited at the edge; no account required.

Added

• Initial public release.

• Gesamtausgabe corpus (GA 1 through GA 354). Rudolf Steiner’s books, lectures, and lecture cycles. Public domain.

• Multilingual retrieval. English translations alongside German originals. Per-edition tracking (translator, year, basis). alignment_confidence flag marks paragraph-count divergence between source and translation.

• MCP server at https://anthroposophy.ai/mcp. Streamable HTTP transport per MCP spec 2025-03-26. Anonymous access; per-IP rate limiting at the edge.

• Tool search_steiner: FTS5 search over the Gesamtausgabe. Porter stemmer with diacritic normalization. Filters: lang, edition, ga, ga_range, from_date, to_date, corpus.

• Tool verify_locus: confirm a GA reference exists in the indexed corpus before citing.

• Tool fetch_passage: retrieve a verbatim Steiner passage by (ga, lecture, passage_number, lang, edition). Returns text and a stable citation URL.

• Tool fetch_parallel: retrieve the same passage across multiple translation editions side-by-side.

• REST API at https://anthroposophy.ai/v1/.

• Browseable passages at /v1/steiner/<lang>/ga/<ga_volume>/<lecture_date>/p<NNN>. Content negotiation returns HTML to browsers, JSON to agents via the .json suffix.