Skip to main content
Glama
theosis-oracc-mcp

theosis-oracc-mcp

by PrattyT85

theosis-oracc-mcp

Read-only MCP server wrapper for the ORACC (Open Richly Annotated Cuneiform Corpus) JSON archive API, built for integration with the Theosis theological research stack.

Features

  • list_projects — List public ORACC projects with optional filtering

  • get_project_metadata — Project config, formats, and witnesses

  • get_project_manifest — Available JSON files for a project

  • list_project_texts — Catalogue entries with designation, language, period, genre

  • get_text — Bounded CDL excerpt with transliteration and translation

  • search_project — Search catalogue by designation, author, or title

Related MCP server: catalogue-api

Setup

uv sync
uv run python -m oracc_mcp.server  # stdio transport

Or add to your Hermes config:

mcp:
  servers:
    oracc-mcp:
      transport: stdio
      command: uv
      args: ["--directory", "/path/to/theosis-oracc-mcp", "run", "python", "-m", "oracc_mcp.server"]

Testing

uv run pytest -v                    # offline tests only
uv run pytest -m live -v           # live tests (ORACC_LIVE=1 required)
ORACC_LIVE=1 uv run pytest -m live -v

ORACC Attribution

This server is a read-only client for the ORACC JSON archive API. projects.json is used for discovery; project metadata, catalogues, corpora, and text editions are read from the current /json/<project-archive>.zip downloads. Archives are bounded and cached only in memory for the lifetime of one client; they are not persisted or redistributed. Results include the archive URL and member path as provenance. ORACC data is provided by the University of Pennsylvania and its project contributors; follow the licence and attribution for the specific source.

Citation: Steve Tinney & Eleanor Robson, 'Oracc JSON Data: A brief introduction for programmers', Oracc: The Open Richly Annotated Cuneiform Corpus, Oracc, 2019 [http://oracc.museum.upenn.edu/doc/opendata/json/]

License

MIT — see LICENSE.

Available Tools

6 tools
get_project_manifestA

Get the JSON manifest listing available files for a project.

Args: project: ORACC project identifier.

Inputs: project (required). Returns: result

get_project_metadataA

Get metadata for an ORACC project (config, formats, witnesses).

Args: project: ORACC project identifier (e.g. 'rimanum' or 'aemw/alalakh/idrimi').

Inputs: project (required). Returns: result

get_textB

Get a bounded excerpt of an ORACC text edition (CDL structure).

Returns source URL, text ID, project, and transliteration/translation text where available. The response is capped to keep output manageable.

Args: project: ORACC project identifier. text_id: ORACC text ID (e.g. 'P295625').

Inputs: project (required), text_id (required). Returns: result

list_projectsA

List ORACC public projects, optionally filtered by name substring.

Args: query: Optional case-insensitive substring filter. limit: Max projects to return (default 50).

Inputs: limit, query. Returns: result

list_project_textsA

List texts in a project's catalogue with key metadata (designation, language, period, genre).

Args: project: ORACC project identifier. query: Optional case-insensitive substring filter on designation. limit: Max texts to return (default 50).

Inputs: project (required), limit, query. Returns: result

search_projectA

Search a project's catalogue for texts matching a query (designation, author, or title).

Args: project: ORACC project identifier. query: Case-insensitive search term. limit: Max results (default 20).

Inputs: query (required), project (required), limit. Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.9/5.0

    Scored across 6 tools

    Disambiguation4/5

    Each tool targets a distinct resource or action, but list_project_texts and search_project both query the project catalogue and could be confused. The descriptions make the browse-versus-search distinction clear, so this is only a minor overlap.

    Naming Consistency5/5

    All tool names follow a consistent verb_noun pattern using snake_case: list, get, and search verbs are applied predictably to projects, metadata, manifests, texts, and project text lists. There are no abrupt style or casing changes.

    Tool Count5/5

    Six tools is a well-scoped set for a read-only ORACC corpus access server. Each tool covers a meaningful step in the workflow from discovering projects to retrieving text content, with no obvious redundancy.

    Completeness4/5

    The tool surface covers project discovery, metadata, file manifests, text listing, text retrieval, and catalogue search, which forms a complete read-only workflow. The main limitation is that get_text only returns a bounded excerpt rather than full text editions, but this appears intentional for output management.

    Maintenance

    ActivitySlowing
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      Provides read-only access to Cordra digital object repositories, enabling AI assistants to explore schemas, retrieve objects, and search using Lucene queries.
      1
      MIT
    • F
      license
      Not graded
      quality
      C
      maintenance
      Provides a read-only API over a book catalogue, offering tools to search books, retrieve book details and series, and inspect provenance and source agreement data.
      -
    • A
      license
      B
      quality
      C
      maintenance
      Enables read-only search and exploration of the GenizahSearch research API, including manuscript search, page browsing, parallel finding, and multi-phrase queries, along with policy resources and research prompts.
      4
      Creative Commons Attribution Non Commercial Share Alike 4.0 International
    • A
      license
      Not graded
      quality
      B
      maintenance
      Provides read-only access to the Internet Archive's Wayback Machine and catalog, enabling retrieval of archived web captures, snapshot metadata, and catalog search.
      BSD 3-Clause