bbaw-dse-mcp
A legacy snapshot of bbaw-dse-mcp: an MCP server exposing Schleiermacher Digital, Praktiken der Monarchie and correspSearch tools for exploratory, conversational research on digital scholarly editions — though the schema lags far behind the documented public-only cutover.
Schleiermacher Digital (
sd_*): full-text document search with doc-type/year/date facets and KWIC snippets, sender/receiver/place/date letter filtering, register search for persons, places and works with optional mention summaries and correspondence stats, plus document reading, passages, diary entries (1808–1811, 1817, 1820–1834) and chronology by day, range or year.Praktiken der Monarchie (
mop_*): full-text search and document retrieval, register/biogramme lookup with family-network extraction, residential topography browsing and filtering by year (1800, 1845, 1872, 1891, 1914), and adjutanten-journal search, entry reading and per-monarch adjutant lists.correspSearch (
cs_*): cross-edition letter search by GND/VIAF person, GeoNames place, Wikidata occupation, dates, gender, role, availability or CMIF URL, with pagination; edition discovery; correspondent-network analysis with top partners.Authority resolution: GND lookups via Lobid, GeoNames place IDs, and Wikidata occupation/entity search to build filters for cross-edition queries.
Legacy/diagnostic surface:
sd_*andmop_*still expose raw XQuery execution, collection browsing, collection stats, raw XML retrieval and database connectivity checks — precisely the private eXist-db features the README says were removed.Not present in this schema: shared tools (
search_documents,search_entities,get_document,get_entity,get_capabilities,get_resource_relations,compare_resource_identities,get_correspondence_statistics), the Jean Paul (jp_*) and edition humboldt digital (ehd_*) modules,authority_*-prefixed tools,_meta.provenanceresponses, and the modern 2026-07-28 protocol surface.Practical caveat: tools assume database accounts and unbounded/legacy response shapes, so the schema likely predates (or misrepresents) the current public, credentials-free server described in the README.
Allows lookup of Wikidata entities and occupations to enrich person and place information.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@bbaw-dse-mcp Find letters from Schleiermacher to his wife in 1809."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
bbaw-dse-mcp
An MCP server for dialogical, exploratory access to digital scholarly editions.
DHd2026 Poster: Agenten im Dienst der Edition
Overview
Instead of complex search forms and navigation structures, this MCP server enables natural-language conversation with digital scholarly editions. An AI agent translates questions like "Who was important for Schleiermacher in 1810?" into structured database queries — and maintains conversational context across follow-up questions.
The current server composes edition backends and reusable external research services into a single MCP endpoint using FastMCP:
Service | Prefix | Backend | Content |
| Public TEI + native search/register pages | Letters, diaries, lectures, chronology, glossary | |
| Public TEI/HTML + GeoJSON/Zotero | Documents on Prussian monarchy and governance | |
| HTTPS connector | Letters, Umfeldbriefe, topics, correspondence networks | |
| Public TEI API v2.3 + native web modules | Letters, travel journals, thematic documents, research, chronology, registers | |
| REST API | Cross-edition correspondence network search | |
Authority services |
| Lobid GND, GeoNames, Wikidata APIs | Reusable identifier and entity lookup |
Related MCP server: WLO MCP Server
Available Tools
The composed server exposes these shared and edition-native tool surfaces:
Surface | Description |
Shared | Cross-edition access, failure-isolated search, correspondence statistics, capabilities, and identity comparison |
Schleiermacher Digital ( | Bounded advanced search, glossary, passages, diaries, chronology, thesaurus, and Bible references |
Praktiken der Monarchie ( | Passages, research literature, biogrammes, adjutant journals, residential topography, court calendar, organigrams, thematic dossiers |
Jean Paul Briefwechsel ( | Native full-text and letter-metadata filters, controlled topics, and curated correspondence networks |
edition humboldt digital ( | TEI catalogs, native search, correspondent/year filters, thematic documents, complete passages, chronology and place-event links |
correspSearch ( | Cross-edition correspondence metadata, normalized relations, typed networks, and edition discovery |
Authority services ( | GND, GeoNames, and Wikidata lookup |
The full signatures and usage boundaries are documented in docs/TOOLS.md; the source-code-derived edition-module inventory is in docs/EDITION_MODULES.md.
Every successful sd_*, mop_*, jp_*, ehd_*, or cs_* response, including
shared calls that resolve edition data, carries MCP _meta.provenance with
consulted sources, configured backends, datasets, and UTC retrieval time.
Humboldt preserves the actual release asserted in returned TEI/canonical
resource URLs (for example v12); HTML searches without release evidence report
release_status="unknown". The other connectors report data_release=null
where no coherent source release exists. API/software versions are never
substituted for data releases.
The capability tool, get_capabilities(check_health=False), reports SD, MoP,
JP, Humboldt, correspSearch, Lobid GND, Wikidata, and GeoNames without exposing
credential values. It declares datasets, modules, filters, releases/interface
versions, licenses, and explicit source limitations. Its default is network-free;
check_health=True runs isolated read-only probes for configured providers. A missing GeoNames
username is visible as configured=false, available=false, and
status="missing_credentials".
search_documents(...) and search_entities(...) return typed pages with
results, offset continuation, per-source counts, failures and filter handling,
rather than bare lists. Both accept strict_filters=False; strict mode rejects
an affected source rather than silently ignoring unsupported filters. Document
search defaults to include_commentary=None (no explicit commentary filter).
Even when every selected backend fails, the structured failure envelope is returned.
total_count is exact only when total_known=true; otherwise it is an
explicit lower bound. results_omitted, extraction_complete, and text
TextExtent.truncated describe different limits; previews carry executable
complete-content routes. SD uses stable source order for repeatable pages.
This is a clean response-contract cutover, not a compatibility shim; see
shared contracts and signatures.
get_entity("sd:…" | "mop:…" | "jp:…" | "ehd:…", include_mentions=True)
now reads the editions' public register pages without a database account.
Edition-specific parsers preserve native metadata, digital text/commentary
mentions, correspondence/authorship, print references and relationships where
exposed. mentions.references is one stable, globally paginated row stream;
mentions.sections retains the full section inventory on every page.
total_count counts rendered rows, not distinct documents or corpus occurrences.
Absent sections and unresolved native destinations remain explicit.
Published inventories are authoritative within their declared source scope:
corpus_complete=true means complete extraction, even when the returned window
has more rows. Unavailable surfaces and actual extraction failures remain
incomplete; published-reference counts are not historical corpus totals.
Shared search, normalized relations and SD/MoP specialist tools now also use
public sources, without a database account or hidden database fallback.
Public retrieval covers every register published in the four editions:
SD's four classic registers plus glossary, Bible and aesthetics thesaurus;
MoP's six classic registers plus bibliography; JP's person/place/two work
registers plus topics and correspondence circles; Humboldt's person/place/
institution/plant/glossary/bibliography/sigle registers. Its drawings gallery
has no published entity IDs and is paged by ehd_list_drawings.
Native types and edition namespaces remain intact. Use public specialist
discovery for SD's extra registers; the manifest separates searchable
entity_types from entity_retrieval_types.
get_resource_relations(ref=...) exposes one common, evidence-backed relation
model for Schleiermacher, MoP, Jean Paul, and Humboldt document resources,
including the explicit Jean-Paul module refs jp:topic:<id> and jp:network:<id>. Its resource kinds
and predicates are closed vocabularies, and every predicate validates the
allowed subject and object kinds. Relations retain provider-scoped references
and external identifiers; the server does not silently merge resources into a
cross-edition sameAs identity. The older get_relations(...) remains
available for its established edition-specific aggregate response shapes.
compare_resource_identities(...) performs that reconciliation explicitly.
It normalizes GND, GeoNames, Wikidata, VIAF, and source URLs, retaining
field-level evidence, conflicts, and partial failures. Only an exact authority
identifier or explicitly equal source URI confirms same_as; equal labels
produce an unconfirmed candidate and never merge records.
Schleiermacher correspondence relations are atomic sender-to-recipient edges
backed by individual letters. MoP relation extraction reads the complete
register-mention set and distinguishes explicit Hofkalendarium participants
from mere mentions. The query-oriented correspSearch equivalent,
cs_search_correspondence_relations(...), retains every sent/received action
with all participants, places, and date assertions from each fetched record.
Breaking correspSearch response migration: cs_search_correspondences
records now expose correspondence data only through letters[].actions[].
Select actions by type (sent or received) and read every participants,
places and dates entry. The singular sender, receiver, send_place,
receive_place and letter-level date* fields have been removed, without
aliases or fallback synthesis. title is display text, not structured evidence.
Tool names and inputs are unchanged; see the migration mapping.
correspSearch metadata searches may start from any supported substantive
filter; the API defaults to ten records but accepts bounded page sizes up to
1,000. Normalized relation responses deliberately process one upstream page
and expose further pages through upstream_continuations.
Jean Paul's native jp_search_documents_advanced and jp_filter_letters
surfaces expose all three letter corpora with exact totals, stable offset
continuation, concrete source collections, and original citation routes. The
first adds full-text, lemma, sender/receiver, year, and counted-facet access;
the second combines correspondence metadata including co-senders,
co-recipients, topics, joint letters, and year bounds. Both fail explicitly on
unexpected source markup. Shared JP work lookup also covers the separate
werke-jp register and retains that collection in search and detail results.
Humboldt uses the conservative ehd namespace (edition humboldt digital).
search_documents(..., editions=["ehd"]), get_document("ehd:H1055854"),
search_entities(..., editions=["ehd"]), and get_entity("ehd:H0012069")
use the same shared adapter/identity surface as the other editions. Native
tools expose exact local TEI-catalog pages and the site's search/filter pages;
ehd_get_document_passages reads complete blocks rather than truncating text.
Use ehd_get_chronology_entry("H0005699") for first-party events and
ehd_get_place_chronology("H0000531") for source-supplied geographic links.
The API's 2.3 interface is distinct from edition v12 (8 June 2026) and
GitHub/Zenodo dataset release identifiers. Entity retrieval primarily uses
public HTML, supplemented by the unversioned ID API where needed because
versioned register XML is defective upstream. Measures such as reaumur
remain public-TEI-retrievable, with references explicitly not_exposed;
production HTML publication has not been established.
Plants use publisher-local selectors (ehd:plant:Abies), bibliography uses
group-qualified Zotero keys (ehd:zotero-667230-R7AXEBAP), and
ehd_list_drawings preserves image crops and source locators without inventing
drawing IDs. Shared date-range/commentary search filters, aggregate networks,
and inferred coordinate or topic graphs are not advertised. See docs/TOOLS.md for
endpoint research, implementation choices, and exact native signatures.
MoP's public full-text search covers Druck, Adjutantenjournale and
Hofkalendarium; semantic letter also includes the published letters in
Druck. Private Korrespondenz/Instruktionen selectors and separate
instruction classification are not established by the public search and are
reported as unsupported, not guessed from titles or returned as false zeros.
mop_get_document_passages provides division/page-filtered text blocks;
shared get_document now returns complete public text or TEI. The current source-gap and acceptance records are
in docs/DATA_SOURCE_GAPS.md and
docs/MOP_ACCEPTANCE.md.
Installation
# Clone the repository
git clone https://github.com/telota/bbaw-dse-mcp.git
cd bbaw-dse-mcp
# Install with uv (recommended)
uv syncRequires Python 3.11+.
Public document retrieval
get_document uses the editions' public TEI exports for sd:, mop:, jp:,
and ehd: references, without database credentials. The internal HTTP receiver
shares bounded downloads and connection handling; edition adapters retain
their own identifiers, catalogs, parsing, and namespaces.
Authority normalization and TEI correspondence-action parsing are shared primitives, not parallel edition implementations. Original URI/XML evidence, multiple participants and uncertain dates are retained; edition adapters still choose their own metadata precedence and resolve their own local IDs. Distinct authority IDs are not silently collapsed, and VIAF/GeoNames are not substituted for GND.
MoP enables optional normal TELOTA Gatekeeper verification: wait for the issued
challenge, submit it to /verify-request, retain the cookie, and retry once.
No browser, forged token, database fallback, or alternate snapshot is used.
Downloads exceeding 16 MiB fail explicitly; XML and document text are not
silently truncated. Facsimile-only documents report availability rather than
automatically generating OCR.
This public-only cutover also covers searches, registers, relations and specialists. Biograms retain family, offices, education and awards; journals retain dated sessions and explicit author/place identities; organigrams retain hierarchies and dated holders. Calendar search uses counted native pages plus local exact-day/text filtering. Catalog-declared calendar UUIDs use public HTML; their unavailable TEI view is an explicit publisher gap.
SD person networks use person-scoped native discovery and letter TEI.
Native whole-corpus correspondent statistics require the publisher's bulk CMIF,
currently unavailable (HTTP 504); this is an error, not an empty network.
get_correspondence_statistics(editions=["sd", "jp", "ehd"]) provides an explicit
correspSearch-backed alternative: live indexed-record counts, or sent/received
counts and bounded partner rankings with ref="sd:S0003676" or person_gnd.
Exact counts remain separate from incomplete partner analysis and corpus coverage.
JP includes print-only An-Briefe metadata. MoP returns unsupported with null
counts because no source association has been verified; native tools stay intact.
AB topography uses the five publisher-linked GeoJSON datasets; literature uses
the live public Zotero group, not a fabricated dated snapshot. Journal text
queries hydrate only their native result window; person/place-only searches
without text/date/monarch narrowing can require a bounded, cold-cache scan.
The obsolete private eXist clients, database/query tools, SD legacy parsers and
exclusive database settings have been removed. Existing .env files must remove
the old SD/AB database URLs, collection paths, usernames/passwords and SD_LOCAL
keys. Use .env.example for the supported public configuration; the GeoNames
account remains an external-service setting. No database compatibility aliases
or hidden fallback are retained.
Configuration
Configuration uses environment variables (prefix EDITIONS_). Create a .env file:
# Schleiermacher Digital public research interfaces; no database account
EDITIONS_SD_PUBLIC_URL=https://schleiermacher-digital.de
# Praktiken der Monarchie public research interfaces; no database account
EDITIONS_AB_PUBLIC_URL=https://actaborussica.bbaw.de
EDITIONS_AB_GATEKEEPER_ENABLED=true
# correspSearch API (no auth required)
EDITIONS_CS_API_URL=https://correspsearch.net/api/v2.0
EDITIONS_CS_SPARQL_URL=https://lod.academy/correspsearch/data/main/sparql
# Required for the GeoNames lookup tools; capability discovery works without it
EDITIONS_GEONAMES_USERNAME=
Usage
Running the server
uv run bbaw-dse-mcpThe default transport remains stdio, including for Claude Desktop. The
server now uses FastMCP 4.0.11+ and the official MCP Python SDK 2.3.0+, which
natively implement the published
MCP 2026-07-28 specification.
Legacy clients still use their initialize handshake on the same server;
modern clients do not initialize or create protocol sessions.
Streamable HTTP deployment
uv run bbaw-dse-mcp --transport http --host 127.0.0.1 --port 8000
# Equivalent transport spelling: --transport streamable-httpThe endpoint is /mcp (--path overrides it). Host and Origin protection is
enabled even when binding to a non-loopback address. Behind a TLS reverse
proxy, explicitly allow the externally visible hostname:
uv run bbaw-dse-mcp --transport http --host 127.0.0.1 --port 8000 \
--allowed-host editions.example.orgRepeat --allowed-host for additional trusted hostnames. --allowed-origin
allows an explicit browser origin; it does not configure CORS. Preserve the
original Host and Origin headers through the proxy, disable buffering for
SSE responses, and configure authentication/access restrictions at your
deployment boundary before exposing the endpoint publicly. The repository
does not configure an OAuth authorization server or issue access tokens.
Modern requests need no worker affinity; legacy stateful HTTP clients still
need session affinity when using multiple workers.
The deprecated HTTP+SSE transport remains available for legacy integrations
with --transport sse (default path /sse). New integrations should use
Streamable HTTP; the modern transport has no standalone GET stream, session
DELETE, or resumable SSE event IDs.
For ASGI embedding, preserve the FastMCP lifespan and the body-primary validation guard:
from bbaw_dse_mcp.main import app
from bbaw_dse_mcp.protocol import modern_http_middleware
http_app = app.http_app(
path="/mcp",
middleware=[modern_http_middleware("/mcp")],
host_origin_protection=True,
allowed_hosts=["editions.example.org"],
)
# Mount http_app with its lifespan, not just its routes.Before release, run uv run python scripts/verify_all_tools.py against the local
composition, or add --url https://your-host/mcp to verify a deployed endpoint.
The release verification guide documents
coverage, semantic checks, expected limitations, JSON reports and failure exit
status. Scoped --tool runs are not full release approval.
Modern protocol and raw smoke requests
Every modern request includes params._meta with
io.modelcontextprotocol/protocolVersion and
io.modelcontextprotocol/clientCapabilities (an empty object is valid).
io.modelcontextprotocol/clientInfo is recommended but optional. Requests
are independently negotiated and can call tools before discovery.
server/discover advertises protocol versions, capabilities, instructions,
and server identity; ordinary results include resultType: "complete".
Cacheable discovery/list/resource results include ttlMs and cacheScope.
Tool diagnostics use standard Python module logging (stderr on stdio), not
the deprecated MCP logging capability. They do not require client logging
opt-in or emit unsolicited protocol log notifications.
On HTTP, include Accept: application/json, text/event-stream,
MCP-Protocol-Version, and Mcp-Method on every request. tools/call,
prompts/get, and resources/read additionally require Mcp-Name matching
the body name or URI. Missing/mismatched routing headers return HTTP 400
with HeaderMismatch (-32020); missing required metadata returns
-32602; unsupported versions return -32022 with supported and
requested. Unknown or removed modern methods (including ping) return
HTTP 404 and -32601. Header-safe values and Base64-sentinel names are
validated by the SDK.
With the HTTP server running, these network-free calls exercise discovery, tool enumeration, and a real capability-tool execution:
META='{"io.modelcontextprotocol/protocolVersion":"2026-07-28","io.modelcontextprotocol/clientCapabilities":{}}'
for METHOD in server/discover tools/list; do
curl -sS http://127.0.0.1:8000/mcp \
-H 'Content-Type: application/json' \
-H 'Accept: application/json, text/event-stream' \
-H 'MCP-Protocol-Version: 2026-07-28' -H "Mcp-Method: $METHOD" \
--data "{\"jsonrpc\":\"2.0\",\"id\":1,\"method\":\"$METHOD\",\"params\":{\"_meta\":$META}}"
done
curl -sS http://127.0.0.1:8000/mcp \
-H 'Content-Type: application/json' \
-H 'Accept: application/json, text/event-stream' \
-H 'MCP-Protocol-Version: 2026-07-28' \
-H 'Mcp-Method: tools/call' -H 'Mcp-Name: get_capabilities' \
--data "{\"jsonrpc\":\"2.0\",\"id\":2,\"method\":\"tools/call\",\"params\":{\"_meta\":$META,\"name\":\"get_capabilities\",\"arguments\":{\"check_health\":false}}}"The default HTTP response is request-scoped SSE, so JSON-RPC messages appear
on data: lines. To check rejection, repeat the last request with
Mcp-Name: wrong_name; expect HTTP 400 and code -32020. To check
negotiation, set both the body version and version header to
2099-01-01; expect HTTP 400 and -32022, not a legacy fallback.
On stdio, write JSON bodies one per line without HTTP headers and keep stdin open until responses arrive (closing it is the graceful shutdown signal):
uv run python - <<'PY'
import asyncio
import json
import sys
async def smoke():
proc = await asyncio.create_subprocess_exec(
sys.executable, "-m", "bbaw_dse_mcp.main",
stdin=asyncio.subprocess.PIPE, stdout=asyncio.subprocess.PIPE,
limit=8 * 1024 * 1024,
)
meta = {
"io.modelcontextprotocol/protocolVersion": "2026-07-28",
"io.modelcontextprotocol/clientCapabilities": {},
}
try:
calls = [
("server/discover", {}),
("tools/list", {}),
("tools/call", {"name": "get_capabilities", "arguments": {}}),
]
for ident, (method, args) in enumerate(calls, 1):
body = {"jsonrpc": "2.0", "id": ident, "method": method,
"params": {"_meta": meta, **args}}
proc.stdin.write((json.dumps(body) + "\n").encode())
await proc.stdin.drain()
reply = json.loads(await asyncio.wait_for(proc.stdout.readline(), 30))
assert reply["id"] == ident and reply["result"]["resultType"] == "complete"
print(json.dumps(reply))
proc.stdin.close()
await asyncio.wait_for(proc.wait(), 10)
assert proc.returncode == 0
finally:
if proc.returncode is None:
proc.kill()
await proc.wait()
asyncio.run(smoke())
PYtests/contract/test_protocol.py covers raw HTTP and subprocess stdio,
per-request errors, cache metadata, streaming-body replay, DNS-rebinding
protection, and actual FastMCP modern/legacy clients. Run with
uv run --extra dev pytest tests/contract/test_protocol.py --no-cov.
Upstream boundaries: SDK v2 supports both protocol eras, MRTR and subscriptions, but core experimental tasks have moved to an optional extension; this server does not register that extension or require client sampling/roots/elicitation. Its edition tools return complete results. A narrow ASGI guard compensates for SDK 2.3.0's header-only era router so modern bodies cannot evade header validation by omitting the version header or supplying a legacy one. It buffers/replays only MCP POST bodies, bounded to 8 MiB and 4,096 chunks (HTTP 413 above those limits); Host/Origin checks remain in FastMCP's outer middleware. See the official SDK v2 changes, FastMCP 4 migration guide, and HTTP transport requirements.
Prompt templates
Clients with MCP prompt support can start common, source-grounded workflows with:
research_correspondence— investigate a person's correspondence network.analyze_edition_document— retrieve and analyze a known SD or MoP document.deep_research— plan and execute a multi-step cross-edition investigation.
These prompts orchestrate the existing MCP tools in the host model. The server does not require a separate model-provider API key.
Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"editions": {
"command": "uv",
"args": ["run", "bbaw-dse-mcp"],
"cwd": "/path/to/bbaw-dse-mcp"
}
}
}Example Conversation
User: Search for letters mentioning Humboldt.
Agent: → search_documents(query="Humboldt", editions=["sd"], doc_types=["letter"])
Inspect the returned native search hits and their source metadata.
User: Who was important for Schleiermacher in 1810?
Agent: → get_correspondence_statistics(editions=["sd"], ref="sd:S0003676",
date_from="1810", date_to="1810", max_letters_to_analyze=500)
Report indexed correspondence counts and the bounded partner ranking;
distinguish frequency from historical importance and index coverage.
User: Can we trace Humboldt's correspondence network beyond this edition?
Agent: → authority_search_gnd(name_query="Humboldt, Wilhelm von")
→ cs_search_correspondent_network(person_gnd="118554727")
Report the analyzed records, authority identities and completeness limits.Tech Stack
FastMCP 4.x / MCP Python SDK 2.x — Native modern and legacy MCP protocol support
httpx — Async public HTTP APIs and pages
lxml — TEI-XML parsing
Pydantic v2 — Data models and settings
Documentation
AGENT.md — Project overview and design decisions
docs/TOOLS.md — Detailed tool specifications
docs/ARCHITECTURE.md — Technical architecture
docs/CODING_GUIDELINES.md — Extension and testing rules
License
MIT
Author
Tim Westphal, BBAW TELOTA
Available Tools
44 toolscs_get_edition_infoA
Informationen über eine Edition in correspSearch abrufen.
PURPOSE: Metadaten zu registrierten Editionen
WHEN TO USE:
User fragt nach verfügbaren Editionen
Für Edition-Discovery
Um die Quelle von gefundenen Briefen zu identifizieren
Args: edition_id: Edition-UUID in correspSearch ctx: FastMCP Context
Returns: EditionInfo mit Metadaten zur Edition
Inputs: edition_id (required). Returns: Information about an edition registered in correspSearch.
cs_get_place_geonames_idA
Get the GeoNames ID for a place (convenience function).
PURPOSE: Quick lookup of a single GeoNames ID for a place.
WHEN TO USE:
When you just need the ID without details
For quick place-to-ID conversion
When you're confident about the place name
WHEN NOT TO USE:
When the place name might be ambiguous → use search_for_geonames_id()
When you need full place details
Args: place_name: Name of the place (e.g., "Berlin", "Paris") country: Optional ISO-2 country code to restrict search (e.g., "DE")
Returns: GeoNames ID as integer, or None if not found
Example: >>> berlin_id = await get_place_geonames_id("Berlin", country="DE") >>> # Returns: 2950159 >>> letters = await search_correspondences(place_geonames=str(berlin_id))
Inputs: place_name (required), country. Returns: result
cs_search_correspondencesA
Cross-Edition Briefsuche über correspSearch API.
PURPOSE: Briefe über Edition-Grenzen hinweg finden
WHEN TO USE:
User möchte Briefe in ALLEN Editionen finden
Cross-Referenzierung zwischen Editionen
Für umfassende Korrespondenz-Netzwerk-Analyse
Suche nach Korrespondenz einer historischen Person
WHEN NOT TO USE:
Für schleiermacher-spezifische Suche → nutze sd_search_letters()
Args: person_gnd: GND-ID(s) - einzelne ID oder Liste (z.B. "118540238" oder ["118540238", "118607626"]) person_viaf: VIAF-ID(s) - einzelne ID oder Liste (alternativ zu GND) start_date: Start-Datum (ISO 8601, z.B. "1810-01-01") end_date: End-Datum (ISO 8601, z.B. "1815-12-31") place_geonames: GeoNames-ID eines Ortes (z.B. "2879139" für Leipzig) occupation_wikidata: Wikidata-ID eines Berufs (z.B. "Q36180" für Schriftsteller) edition_id: UUID einer Edition zur Filterung cmif_url: URL einer CMIF-Datei (z.B. "https://gams.uni-graz.at/context:hsa/CMIF") availability: "online", "print" oder "hybrid" text_query: Volltextsuche in Briefinhalten (experimentell, undokumentiert) gender: "male" (männlich), "female" (weiblich), oder "unknown" (unbekannt) role: "sent" (nur als Absender), "received" (nur als Empfänger), oder "mentioned" (nur erwähnt) place_role: "sent" (Schreibort) oder "received" (Empfangsort) page: Seite der Ergebnisse (1-indiziert, je 100 Treffer) max_results: Maximale Ergebnisse (zur Anzeige-Begrenzung) ctx: FastMCP Context
Returns: CorrespSearchResult mit Briefen und Paginierungs-Info
Note: Mehrere Personen (Liste) werden mit AND kombiniert - findet nur Briefe, die ALLE angegebenen Personen enthalten.
Inputs: page, role, gender, cmif_url, +11 more. Returns: Result set from correspSearch API with pagination info.
cs_search_correspondent_networkA
Korrespondenz-Netzwerk einer Person analysieren.
PURPOSE: Netzwerkanalyse von Korrespondenzen
WHEN TO USE:
User fragt "Mit wem korrespondierte Person X?"
Netzwerk-Visualisierung vorbereiten
Wichtigste Korrespondenzpartner identifizieren
Args: person_gnd: GND-ID der fokalen Person start_date: Optional: Start-Datum für Zeitfilter end_date: Optional: End-Datum für Zeitfilter max_correspondents: Maximale Anzahl Korrespondenten im Ergebnis max_letters_to_analyze: Maximale Anzahl Briefe zu analysieren (pro Richtung) ctx: FastMCP Context
Returns: Dict mit Netzwerk-Statistiken und Top-Korrespondenten
Inputs: person_gnd (required), end_date, start_date, max_correspondents, +1 more. Returns: object
cs_search_for_geonames_idA
Search for GeoNames IDs by place name.
PURPOSE: Find GeoNames identifiers for places to use as filters.
WHEN TO USE:
When you need a GeoNames ID for correspSearch place filtering
To resolve place names to standard identifiers
For geographic analysis of correspondence networks
WHEN NOT TO USE:
For person or organization names → use search_for_gnd_id()
When you already have a GeoNames ID
Args: place_query: Place name to search for (e.g., "Berlin", "Paris") country: Optional ISO-2 country code to restrict search (e.g., "DE", "FR") limit: Maximum number of results to return
Returns: List of dicts with place information: - geonameId: The GeoNames ID (use this for correspSearch) - name: Place name - countryCode: ISO-2 country code - lat, lng: Coordinates - adminName1: First-level admin division (e.g., state) - population: Population count (if available)
Example: >>> results = await search_for_geonames_id("Berlin", country="DE") >>> geonames_id = str(results[0]['geonameId']) >>> letters = await search_correspondences(place_geonames=geonames_id)
Inputs: place_query (required), limit, country. Returns: result
cs_search_for_gnd_idA
Search for GND IDs by name using Lobid GND API.
PURPOSE: Find GND identifiers for persons, places, or corporate bodies.
WHEN TO USE:
When you need a GND ID for correspSearch filtering
To resolve person names to standard identifiers
For enriching metadata with authority data
WHEN NOT TO USE:
For full entity data → use get_gnd_entity() tools
For non-GND authority data → use other services
Args: name_query: Name or term to search for (e.g., "Goethe", "Humboldt") limit: Maximum number of results to return
Returns: List of dicts with 'id' (GND URI) and 'label' for matching entities
Example: >>> results = await search_for_gnd_id("Wilhelm von Humboldt") >>> gnd_id = results[0]['id'].split('/')[-1] # Extract ID from URI >>> letters = await search_correspondences(person_gnd=gnd_id)
Inputs: name_query (required), limit. Returns: result
cs_search_for_wikidata_occupationA
Search for occupation entities in Wikidata.
PURPOSE: Find Wikidata IDs for occupations to filter correspondence by profession.
WHEN TO USE:
When you need a Wikidata ID for correspSearch occupation filtering
To find letters by correspondent profession (writers, philosophers, etc.)
For professional network analysis
WHEN NOT TO USE:
For person names → use search_for_gnd_id()
For places → use search_for_geonames_id()
For non-occupation entities → use search_wikidata_entity()
Args: occupation_query: Occupation name to search for (e.g., "Dichter", "Philosoph", "Maler") limit: Maximum number of results to return
Returns: List of dicts with occupation entities: - id: Wikidata ID (e.g., "Q36180") - label: Occupation name in German - description: Brief description - uri: Full Wikidata URI
Example: >>> results = await search_for_wikidata_occupation("Schriftsteller") >>> occupation_id = results[0]['id'] # e.g., "Q36180" >>> letters = await search_correspondences(occupation_wikidata=occupation_id)
Inputs: occupation_query (required), limit. Returns: result
cs_search_wikidata_entityA
Search for any entity in Wikidata.
PURPOSE: General-purpose Wikidata entity search for various use cases.
WHEN TO USE:
When searching for entities that aren't occupations
For broader Wikidata lookups
When you need flexibility in entity type
WHEN NOT TO USE:
Specifically for occupations → use search_for_wikidata_occupation()
For GND/GeoNames data → use respective tools
Args: query: Search term for any Wikidata entity limit: Maximum number of results to return language: Language code for labels (default: "de")
Returns: List of dicts with entity information: - id: Wikidata ID (e.g., "Q36180") - label: Entity name - description: Brief description - uri: Full Wikidata URI
Example: >>> results = await search_wikidata_entity("Aufklärung") >>> entity_id = results[0]['id']
Inputs: query (required), limit, language. Returns: result
mop_browse_documentsA
Browse files and subcollections in MoP.
PURPOSE: Überblick über verfügbare Dateien in der MoP-Edition
WHEN TO USE:
User möchte sehen, was in der Edition verfügbar ist
Exploration ohne konkreten Suchbegriff
WHEN NOT TO USE:
Bei konkreter Suche → nutze search_documents()
Für Metadaten → nutze get_document() danach
Args: collection: Collection-Name (Texte, Register) limit: Maximale Anzahl Dateien ctx: FastMCP Context für Progress
Returns: Dict mit 'files' und 'subcollections' Liste
Inputs: limit, collection. Returns: object
mop_check_database_connectionA
Check if the database is reachable and responsive.
PURPOSE: Verify database connectivity for troubleshooting.
WHEN TO USE:
When other tools fail unexpectedly
To verify setup is working
Health monitoring
Returns: DatabaseStatus object with connection status, version, and paths
Inputs: no named parameters. Returns: object
mop_execute_xqueryA
Execute a raw XQuery against the database.
PURPOSE: Run custom queries for advanced users or debugging.
WHEN TO USE:
Other tools don't provide the needed functionality
Debugging or exploring data structure
Complex custom queries
WHEN NOT TO USE:
For common operations, use specific tools instead
Don't use for write operations (read-only!)
Args: query: XQuery string to execute max_results: Maximum number of results to return
Returns: Raw query result as string (usually XML)
Inputs: query (required), max_results. Returns: result
mop_extract_family_networkA
Familiennetzwerk aus einem Biogramm extrahieren.
PURPOSE: Verwandtschaftsbeziehungen analysieren
WHEN TO USE:
Für genealogische Forschung
Um höfische Netzwerke zu rekonstruieren
Analyse von Familiendynastien am Hof
Args: biogramm_id: XML-ID des Biogramms ctx: FastMCP Context
Returns: Dict mit Familienrelationen strukturiert nach Typ
Inputs: biogramm_id (required). Returns: object
mop_get_adjutanten_journal_entryA
Retrieve full text and metadata of a specific journal entry.
PURPOSE: Get complete details of a daily court journal entry.
WHEN TO USE:
After finding entries with search_adjutanten_journals()
User wants to read the full journal entry for a specific day
User needs detailed information about activities on a specific date
WHEN NOT TO USE:
For searching multiple entries → use search_adjutanten_journals()
For biographical data → use get_register_entry()
Args: document_id: Document ID from search results (e.g., "P0005285") ctx: FastMCP Context for progress reporting
Returns: Dict with: - id: Document ID - monarch: Which monarch's reign - date_from/date_to: Time range covered - shelfmark: Archive reference - days: List of daily entries, each with: - date: ISO date - place: Location - authors: Adjutants on duty - text: Full journal text for that day - url: Link to online edition
Raises: ToolError: If document not found or retrieval fails
Inputs: document_id (required). Returns: object
mop_get_biogramm_by_idA
Vollständiges Biogramm mit allen Details abrufen.
PURPOSE: Detaillierte biografische Daten einer Person abrufen
WHEN TO USE:
Nach search_biogramme() um Details zu bekommen
Um Familiennetzwerk, Karriere, Besitz zu analysieren
Für vollständige prosopographische Information
Args: biogramm_id: XML-ID des Biogramms (z.B. "P0005251") ctx: FastMCP Context
Returns: Dict mit allen biografischen Daten strukturiert
Inputs: biogramm_id (required). Returns: object
mop_get_documentA
Vollständiges Dokument abrufen.
PURPOSE: Detaillierte Ansicht eines spezifischen Dokuments
WHEN TO USE:
User möchte ein Aktenstück lesen
Nach erfolgreicher Suche → Details anzeigen
WHEN NOT TO USE:
Für Übersicht → nutze browse_documents() oder search_documents()
Args: document_id: Die xml:id des Dokuments include_xml: Ob TEI-XML inkludiert werden soll ctx: FastMCP Context
Returns: Document-Objekt mit Metadaten und Content
Inputs: document_id (required), include_xml (required). Returns: Generic document (letter, diary, lecture, etc.).
mop_get_register_entryA
Detailansicht eines Registereintrags.
PURPOSE: Vollständige Informationen zu Person, Ort, etc.
WHEN TO USE:
Nach Registersuche für Details
Für biographische/geographische Informationen
Args: entry_id: ID des Registereintrags register_type: Register-Typ ctx: FastMCP Context
Returns: Dict mit allen verfügbaren Informationen
Inputs: entry_id (required), register_type. Returns: object
mop_get_residential_topographyA
Fetch complete residential topography dataset for a specific year.
PURPOSE: Retrieve GeoJSON data with addresses and locations of court officials, royal family members, and other persons/institutions in Berlin and Prussia.
WHEN TO USE:
User asks about where people lived in a specific year
User wants to analyze residential patterns
User needs geographic distribution of court members
For mapping and spatial analysis
WHEN NOT TO USE:
For biographical register data → use search_register()
For document texts → use browse_documents() or search_documents()
Args: year: Year for which to retrieve data (1800, 1845, 1872, 1891, or 1914) ctx: FastMCP Context for progress reporting
Returns: Dict with: - year: The requested year - total_features: Total number of entries - features_with_coordinates: Number of entries with valid geographic coordinates - categories: Count of entries per category - cities: Count of entries per city - sample_features: First 10 features as examples - query_methods: Available search methods
Raises: ToolError: If year is not available or API request fails
Inputs: year (required). Returns: object
mop_list_adjutanten_by_monarchA
List all adjutants who served under a specific monarch.
PURPOSE: Identify who served as adjutant and when.
WHEN TO USE:
User asks "Who served as adjutant under Wilhelm I?"
User wants to know the rotation of adjutants
User researches prosopography of court officials
WHEN NOT TO USE:
For full biographical data → use search_register()
For journal content → use search_adjutanten_journals()
Args: monarch: Monarch name (Friedrich_Wilhelm_IV, Wilhelm_I, Wilhelm_II, Friedrich_III) ctx: FastMCP Context for progress reporting
Returns: Dict with: - monarch: Monarch name - total_entries: Number of journal entries analyzed - adjutants: List of adjutants with: - name: Full name - person_key: Register ID - entries_count: How many times they wrote entries - date_range: First and last appearance
Raises: ToolError: If monarch is invalid or query fails
Inputs: monarch (required). Returns: object
mop_list_available_wohntopo_yearsA
List all available years for residential topography data.
PURPOSE: Show which years have Wohntopographie datasets available.
WHEN TO USE:
User asks about available time periods
User wants to know which years can be queried
Before choosing a year for analysis
Returns: Dict with available years and description
Inputs: no named parameters. Returns: object
mop_search_adjutanten_journalsA
Search in Adjutantenjournale (court adjutant journals).
PURPOSE: Find daily court journal entries documenting royal activities, audiences, meetings, and events. Each entry shows who was on duty and what happened.
WHEN TO USE:
User asks "What did the king do on [date]?"
User wants to know about daily court life
User searches for specific events, persons, or places mentioned in journals
User wants to track activities of a specific monarch
User researches who served as adjutant when
WHEN NOT TO USE:
For biographical data → use search_register()
For correspondence → use letters search in Schleiermacher Digital
For institutional documents → use browse_documents()
Args: query: Full-text search term (searches in journal text) monarch: Filter by monarch (Friedrich_Wilhelm_IV, Wilhelm_I, Wilhelm_II, Friedrich_III) date_from: Start date in ISO format (YYYY-MM-DD), e.g., "1861-01-01" date_to: End date in ISO format (YYYY-MM-DD), e.g., "1861-12-31" person_key: Filter by person mentioned (register key, e.g., "P0002157") place_key: Filter by place mentioned (register key, e.g., "P0003556") limit: Maximum number of results (default: 50) ctx: FastMCP Context for progress reporting
Returns: List of journal entries with: - id: Document ID - monarch: Which monarch's reign - date_from/date_to: Time range covered - place: Where the court was located - authors: Adjutants who wrote the entry - snippet: Text excerpt showing matched content - url: Link to full entry on website
Raises: ToolError: If monarch is invalid or query fails
Inputs: limit, query, date_to, monarch, +3 more. Returns: result
mop_search_biogrammeA
Suche in MoP-Biogrammen (detaillierte Biografien).
PURPOSE: Detaillierte biografische Einträge finden
WHEN TO USE:
User sucht nach Person für biografische Details
Um Familienverhältnisse, Karriere, Besitztümer zu recherchieren
Prosopographische Forschung zu Hofbeamten
WHEN NOT TO USE:
Für einfache Registersuche → nutze search_register("personen")
Für Volltextsuche in Dokumenten → nutze search_documents()
Args: query: Suchbegriff (Name) birth_year: Filter nach Geburtsjahr death_year: Filter nach Sterbejahr max_results: Maximale Ergebnisse ctx: FastMCP Context
Returns: Liste von Biogramm-Treffern mit id, name, birth, death, gnd
Inputs: query (required), birth_year, death_year, max_results. Returns: result
mop_search_documentsA
Volltextsuche in MoP-Dokumenten.
PURPOSE: Dokumente finden, die einen bestimmten Begriff enthalten
WHEN TO USE:
User sucht nach Person, Institution, Thema
Explorative Suche zu höfischen Praktiken
WHEN NOT TO USE:
Für strukturierte Registersuche → nutze search_register()
Args: keyword: Suchbegriff collection: Collection (Texte) max_results: Maximale Ergebnisse ctx: FastMCP Context
Returns: Liste von SearchResult-Objekten
Inputs: keyword (required), collection, max_results. Returns: result
mop_search_registerA
Suche in MoP-Registern.
PURPOSE: Strukturierte Registereinträge finden
WHEN TO USE:
User sucht nach Person, Ort, Institution, Hof
Um IDs für weitere Suchen zu bekommen
WHEN NOT TO USE:
Für Volltextsuche → nutze search_documents()
Args: query: Suchbegriff register_type: Register (personen, orte, institutionen, hoefe, werke, aemter) max_results: Maximale Ergebnisse ctx: FastMCP Context
Returns: Liste von Register-Einträgen mit id, name, type, gnd (optional)
Inputs: query (required), max_results, register_type. Returns: result
mop_search_residential_topographyA
Search residential topography data with multiple filters.
PURPOSE: Find persons/institutions by various criteria in historical address data.
WHEN TO USE:
User searches for specific person by name
User wants to know who lived at a specific address
User wants to filter by profession, category, or location
User needs to find all court members in a certain category
WHEN NOT TO USE:
For biographical details → use get_register_entry()
For full dataset overview → use get_residential_topography()
Args: year: Year for data (1800, 1845, 1872, 1891, or 1914) name: Last name (partial match, case-insensitive) vorname: First name (partial match, case-insensitive) kategorie: Category filter (e.g., "Königliche Familie", "Diplomatie") taetigkeit: Occupation/activity filter stadt: City filter (partial match) strasse: Street name filter (partial match) ediarum_id: Exact Ediarum-ID match only_with_coordinates: Return only entries with valid geographic coordinates max_results: Maximum number of results to return ctx: FastMCP Context
Returns: Dict with: - year: The queried year - filters_applied: List of active filters - total_matches: Number of matching entries - returned_results: Number of results in response (limited by max_results) - results: List of matching features
Raises: ToolError: If year is not available or no filters provided
Inputs: year (required), max_results (required), only_with_coordinates (required), name, +6 more. Returns: object
sd_check_database_connectionA
Check if the database is reachable and responsive.
PURPOSE: Verify database connectivity for troubleshooting.
WHEN TO USE:
When other tools fail unexpectedly
To verify setup is working
Health monitoring
Returns: DatabaseStatus object with connection status, version, and paths
Inputs: no named parameters. Returns: Database connection status.
sd_execute_xqueryA
Execute a raw XQuery against the database.
PURPOSE: Run custom queries for advanced users or debugging.
WHEN TO USE:
Other tools don't provide the needed functionality
Debugging or exploring data structure
Complex custom queries
WHEN NOT TO USE:
For common operations, use specific tools instead
Don't use for write operations (read-only!)
Args: query: XQuery string to execute max_results: Maximum number of results to return
Returns: Raw query result as string (usually XML)
Inputs: query (required), max_results. Returns: result
sd_filter_lettersA
Filter letters by sender, receiver, place, and time period.
PURPOSE: Filter letters by specific criteria (FAST - uses cache)
WHEN TO USE:
User asks "letters from/to person X"
User asks "letters from Berlin"
User asks "letters between 1810 and 1815"
For correspondence network analyses
WHEN NOT TO USE:
For keyword search → use search_by_keyword()
For register search → use search_register()
Args: sender: Person ID of sender (from register, e.g., "S0003676") receiver: Person ID of receiver (from register, e.g., "S0003677") send_place: Place ID or name of sending location (e.g., "S0000065" or "Berlin") not_before: Earliest date (ISO 8601, e.g., "1810-01-01") not_after: Latest date (ISO 8601, e.g., "1815-12-31") max_results: Maximum results ctx: FastMCP Context
Returns: List of Letter objects
Inputs: sender, receiver, not_after, not_before, +2 more. Returns: result
sd_get_chronology_entriesA
Retrieve chronology entries for a date range.
PURPOSE: Access events from Schleiermacher's life across a time period
WHEN TO USE:
User asks what happened during a specific period
User wants biographical timeline for a date range
For temporal analysis of Schleiermacher's life events
WHEN NOT TO USE:
For single date → use get_chronology_entry()
For entire year → use get_chronology_year()
For keyword search → use search_documents()
Args: date_from: Start date in ISO 8601 format (YYYY-MM-DD) date_to: End date in ISO 8601 format (YYYY-MM-DD) ctx: FastMCP Context
Returns: List of dictionaries with date info and event descriptions, sorted chronologically. Includes both specific dates and date ranges that overlap with the query range.
Inputs: date_to (required), date_from (required). Returns: result
sd_get_chronology_entryA
Retrieve chronology entries for a specific date.
PURPOSE: Access events from Schleiermacher's life on a specific date
WHEN TO USE:
User asks what happened on a specific date
User wants biographical information for a particular day
After search → get full chronology entry
WHEN NOT TO USE:
For date range → use get_chronology_entries()
For entire year → use get_chronology_year()
For keyword search → use search_documents()
Args: date: Date in ISO 8601 format (YYYY-MM-DD), e.g., "1785-08-29" ctx: FastMCP Context
Returns: List of dictionaries with date info and event descriptions. Multiple events may occur on the same date.
Inputs: date (required). Returns: result
sd_get_chronology_yearA
Retrieve all chronology entries for a specific year.
PURPOSE: Access complete biographical timeline for a year in Schleiermacher's life
WHEN TO USE:
User asks "What happened in 1785?"
User wants overview of activities in a specific year
For annual biographical summaries
WHEN NOT TO USE:
For specific date → use get_chronology_entry()
For date range spanning multiple years → use get_chronology_entries()
For keyword search → use search_documents()
Args: year: Year (e.g., 1785) ctx: FastMCP Context
Returns: Dictionary with year, heading, and list of chronology entries. Entries are sorted chronologically within the year.
Inputs: year (required). Returns: object
sd_get_collection_statsA
Get statistics about documents in a collection.
PURPOSE: Understand the size and content of a collection.
WHEN TO USE:
User asks "how many documents are there?"
Getting an overview before browsing/searching
Verifying data availability
Args: collection: Collection path (relative to data_path). Empty = root.
Returns: CollectionStats object with document counts and metadata
Inputs: collection. Returns: Statistics about a collection.
sd_get_correspondent_statsA
Statistics about correspondents (for network analysis).
PURPOSE: Overview of most important correspondents
WHEN TO USE:
User asks "Who was important for X?"
User asks "Most frequent correspondents"
For quantitative analyses
Args: year: Optional year filter min_letters: Minimum number of letters ctx: FastMCP Context
Returns: List of dicts with person_name, total, letters_sent, letters_received
Inputs: year, min_letters. Returns: result
sd_get_diary_entriesA
Retrieve diary entries for a date range.
PURPOSE: Access multiple diary entries across a time period
Available years: 1808-1811, 1817, 1820-1834. Note: 1812-1816 and 1818-1819 are not extant.
WHEN TO USE:
User asks for diary entries in a date range
User wants to see activities over a period
For temporal analysis of diary content
WHEN NOT TO USE:
For single date → use get_diary_entry()
For keyword search → use search_in_documents()
Args: date_from: Start date in ISO 8601 format (YYYY-MM-DD) date_to: End date in ISO 8601 format (YYYY-MM-DD) ctx: FastMCP Context
Returns: List of dictionaries, each with date, left_side, right_side content
Inputs: date_to (required), date_from (required). Returns: result
sd_get_diary_entryA
Retrieve a specific diary entry by date.
PURPOSE: Access a specific day's entry from Schleiermacher's diary
Available years: 1808-1811, 1817, 1820-1834. Note: 1812-1816 and 1818-1819 are not extant.
WHEN TO USE:
User asks for diary entry on specific date
User wants to know what happened on a particular day
After search → get full diary entry
WHEN NOT TO USE:
For date range → use get_diary_entries()
For keyword search → use search_in_documents()
Args: date: Date in ISO 8601 format (YYYY-MM-DD), e.g., "1808-01-01" ctx: FastMCP Context
Returns: Dictionary with date, content from left side, content from right side, and raw XML of the entry
Inputs: date (required). Returns: object
sd_get_document_by_idA
Retrieve complete document.
PURPOSE: Detailed view of a specific document as Markdown
WHEN TO USE:
User wants to read a letter/diary entry
After successful search → display details
For citations and text analysis
WHEN NOT TO USE:
For browsing/exploring → use list_collections() or list_collection_contents()
For keyword search → use search_by_keyword()
Args: document_id: The xml:id of the document ctx: FastMCP Context
Returns: Formatted markdown string with document content
Inputs: document_id (required). Returns: result
sd_get_document_passagesA
Retrieve text passages from a specific document.
PURPOSE: Get actual text content from a document found via search
WHEN TO USE:
After search returns relevant document IDs
User wants to read specific sections
Need actual quotes for citations
WHEN NOT TO USE:
For discovery → use search_documents_direct()
For full document → use get_document_by_id()
Args: document_id: The xml:id of the document query: Optional search term to highlight/filter passages division: Filter to specific div by @n value page: Filter to specific page by pb/@n context_size: Characters of context around matches max_passages: Maximum passages to return ctx: FastMCP Context
Returns: List of Passage objects with text and location info
Inputs: document_id (required), page, query, division, +2 more. Returns: result
sd_get_file_infoA
Get basic metadata for a single document file.
PURPOSE: Extract basic TEI metadata from a document without full parsing.
WHEN TO USE:
After browsing, to get details about specific files
When you need title, date, or ID information
For quick metadata extraction
WHEN NOT TO USE:
For full document content → use get_document tools
For multiple files at once → use browse then call this for each
Args: file_path: Full path to the file (relative to /db) or just filename ctx: FastMCP Context for progress reporting
Returns: Dict with 'id', 'title', 'date', 'path', 'mime_type', 'size_bytes', 'modified'
Inputs: file_path (required). Returns: Metadata information about a file.
sd_get_raw_document_by_idA
Retrieve raw XML document by its xml:id.
PURPOSE: Fetch raw XML for a document using its TEI xml:id attribute.
WHEN TO USE:
After finding a document via search or browse
When you need the raw XML for custom processing
As a low-level building block for edition-specific tools
WHEN NOT TO USE:
For exploring/discovering documents → use browse or search tools
When you only have a file path → use get_raw_document_by_path()
Args: document_id: The xml:id attribute of the TEI document collection: Optional collection filter (relative to data_path) ctx: FastMCP Context for progress reporting
Returns: RawDocument with id and xml content
Inputs: document_id (required), collection. Returns: Raw XML document from the database (unparsed).
sd_get_raw_document_by_pathA
Retrieve raw XML document by its file path.
PURPOSE: Fetch raw XML for a document using its database path.
WHEN TO USE:
When you have a file path from browse_collection
When accessing documents with known paths
Faster than ID-based lookup when path is known
WHEN NOT TO USE:
When you have an xml:id → use get_raw_document_by_id()
For exploring documents → use browse or search tools
Args: doc_path: Path to the document (relative to /db) ctx: FastMCP Context for progress reporting
Returns: RawDocument with id, xml content, and path
Inputs: doc_path (required). Returns: Raw XML document from the database (unparsed).
sd_get_register_entryA
Get detailed information about a register entry.
PURPOSE: Complete register entry details, optionally with mentions
WHEN TO USE:
After register search for details about a person/place/work
For biographical information
To find mentions in letters/diaries (set include_mentions=True)
WHEN NOT TO USE:
For document details → use get_document_by_id()
Args: entry_id: xml:id of the register entry (e.g., "S0003676") include_mentions: If True, fetch mentions across letters, diaries, lectures max_mentions: Max mentions per category when include_mentions=True (default 20) ctx: FastMCP Context
Returns: PersonEntry, PlaceEntry, WorkEntry, or dict based on entry type
Inputs: entry_id (required), max_mentions, include_mentions. Returns: result
sd_list_collection_contentsA
List files and subcollections in a collection.
PURPOSE: List files and subcollections without parsing document contents.
WHEN TO USE:
User wants to see what's available in a collection
Exploring collection structure
Fast overview of files and folders
WHEN NOT TO USE:
For document metadata → use get_file_info() afterward
For specific keyword search → use edition-specific search tools
For full document details → use get_document tools
Args: collection: Collection name/path (relative to data_path). Empty = root. limit: Maximum number of files to return ctx: FastMCP Context for progress reporting
Returns: Dict with 'collection_path', 'file_count', 'files', and 'subcollections'
Inputs: limit, collection. Returns: Contents of a collection (files and subcollections).
sd_list_collectionsA
List available sub-collections in the database.
PURPOSE: Explore the database structure and find available collections.
WHEN TO USE:
User wants to know what data is available
Exploring the database structure
Finding the correct collection name for other queries
Args: parent: Parent collection path (relative to data_path). Empty = root.
Returns: Collection object with path, collections list, and document_count
Inputs: parent. Returns: Information about an eXist-db Collection.
sd_search_documentsA
Search documents in Schleiermacher Digital via eXist-db with facets.
PURPOSE: Primary search tool for finding documents by text, type, year, etc.
WHEN TO USE:
Full-text search in documents
Filter by document type (letter, diary, lecture, etc.)
Filter by year or date range
Any search query for document discovery
WHEN NOT TO USE:
For structured letter filtering by sender/receiver → use filter_letters()
For register search (persons, places) → use search_register()
For detailed text passages → use get_document_passages()
⚠️ CITATION WARNING: Only cite documents returned by this tool. Each result includes:
document_id: Use this EXACT value for citations
citation_url: Use this EXACT URL, do not construct your own NEVER invent document IDs or URLs that were not returned in the results.
Valid doc_types (from index):
"letter" or "letter fs" (8071 documents)
"lecture" or "lecture fs" (143 documents)
"chronology" (61 documents)
"intro" (42 documents)
"chronology-intro" (29 documents)
"diary" or "diary fs" (21 documents)
Args: query: Search terms (space-separated for multiple terms) doc_types: Filter by document types (letter, letter fs, diary, diary fs, lecture, lecture fs, chronology, intro) years: Filter by years (e.g., ["1810", "1811"]) date_from: Earliest date (ISO 8601, e.g., "1810-01-01") date_to: Latest date (ISO 8601, e.g., "1815-12-31") include_commentary: If True, search in commentary too (default: True) use_or_logic: If True, use OR between terms (default); if False, use AND limit: Maximum results (default: 50) ctx: FastMCP Context
Returns: List of SearchResult objects with document_id, title, type, date, kwic_snippets, and citation_url
Inputs: limit (required), query (required), use_or_logic (required), include_commentary (required), +4 more. Returns: result
sd_search_registerA
Search register using Lucene fulltext index.
PURPOSE: Find persons, places, works, or organizations using indexed search
WHEN TO USE:
User searches for person, place, work, or organization
For biographical/geographical/bibliographical information
To get IDs for further searches
WHEN NOT TO USE:
For fulltext search in letters/diaries → use search_by_keyword()
Args: query: Search term (name, title, etc.) register_type: Optional filter - 'person', 'place', 'work', 'org', or None for all max_results: Maximum results ctx: FastMCP Context
Returns: List of register entry dicts with id, title, desc, type
Inputs: query (required), max_results, register_type. Returns: result
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 44 tools
There are multiple pairs of essentially identical tools (sd_execute_xquery/mop_execute_xquery, sd_check_database_connection/mop_check_database_connection) and very similar search tools across editions (sd_search_documents/mop_search_documents). Additionally, cs_search_for_geonames_id and cs_get_place_geonames_id are confusingly similar, making tool selection difficult for an agent.
The naming generally follows a consistent prefix_verb_noun pattern (e.g., sd_search_documents, mop_get_register_entry, cs_search_correspondences). Minor deviations exist: 'wohntopo' in mop_list_available_wohntopo_years contrasts with 'residential_topography' in other mop tools, and cs_get_place_geonames_id uses 'get' instead of 'search_for' seen in similar tools.
44 tools is a heavy count for a single MCP server. While the server covers three distinct edition projects (Schleiermacher Digital, MoP, correspSearch), the count is inflated by redundant utilities (duplicate execute_xquery and check_database_connection) and many specific niche tools. The count is borderline but not extremely excessive given the multi-edition scope.
The tool set provides broad read-only coverage for the three digital editions, including browsing, full-text search, register lookups, document retrieval, and specialized features like diary/chronology access, biographical searches, residential topography, and cross-edition correspondence search. Minor gaps exist such as no direct get-by-ID for residential topography entries and no tool to retrieve a specific correspSearch letter, but overall the domain is well covered.
Maintenance
Related MCP Connectors
MCP server for searching Airweave collections with natural language queries.
Scholarly search: OpenAlex, Crossref, arXiv, OpenCitations and PubMed in one endpoint.
Multi-engine scholarly research server for search, traversal, full text, and reading lists.
Agentic search over your Dewey document collections from any MCP-compatible client.
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceEnables searching OpenAleph entities and documents using natural language queries through the MCP protocol.13 npm1MIT
- AlicenseAqualityDmaintenanceMCP server for WirLernenOnline.de that enables searching and retrieving educational materials, collections, topic pages, and metadata through natural language, compatible with OpenAI and Claude.12Apache 2.0
- FlicenseNot gradedqualityCmaintenanceEnables querying the Historisches Grundbuch Basel corpus, including full-text search, person lookups, and property dossier retrieval, through MCP-compatible clients like Claude.-
- AlicenseNot gradedqualityCmaintenanceA read-only MCP server that enables natural language querying of DSpace 7+ repositories via the REST API, allowing users to search, retrieve items, and analyze repository data.38 PyPI3MIT