canonical: https://jentic.com/apis/bethmardutho.org/bethmardutho # Bethmardutho SEDRA IV API Jentic publishes the only available OpenAPI specification for SEDRA IV API, keeping it validated and agent-ready. The SEDRA IV API exposes the Syriac Electronic Data Research Archive - a linguistic and literary database of Syriac lexicons with grammatical and morphological information for each entry. It supports word lookup by SEDRA word id or by Syriac text in consonantal, partially vocalised, or fully vocalised forms, and lexeme lookup by id. The spec is small but specialised, with two read-only endpoints serving JSON for academic and digital humanities applications. ## For AI agents Look up Syriac words and lexemes by SEDRA id or vocalised text in the SEDRA IV linguistic database. ## Scope Does not handle translation, audio transliteration, or full-text manuscript search - use for Syriac word and lexeme lookups in the SEDRA database only. ## Capabilities - Look up a Syriac word by SEDRA word id - Look up a Syriac word by fully, partially, or non-vocalised Syriac text - Retrieve a lexeme entry with grammatical metadata by lexeme id - Cross-reference word forms to their lexeme parents - Use CORS-enabled JSON responses directly from a browser-based reading tool ## Use cases ### Syriac Reading Tool Build a Syriac reading or learning tool that resolves clicked words to their full SEDRA entries, returning grammatical information, root, and lexeme. The `/word/{id}` endpoint accepts both SEDRA numeric ids and Syriac text in any vocalisation, while `/lexeme/{id}` returns the underlying lexeme. CORS is enabled, so a static page can call the API directly without a backend. Example prompt: Given the clicked Syriac word ܐܰܒܳܪܳܐ, GET /word/ܐܰܒܳܪܳܐ.json and render the lexeme reference plus grammatical fields ### Lexicon Cross-Reference Cross-reference a corpus of Syriac texts against the SEDRA lexicons to produce a frequency or concordance list. Each unique word form can be resolved through `/word/{id}` and grouped by the lexeme returned through `/lexeme/{id}.` Building a corpus enrichment pipeline is typically half a day to a day depending on input size and caching strategy. Example prompt: For each word form in a Syriac text, GET `/word/{form}.json` and aggregate the lexeme ids to produce a frequency list ### Digital Humanities Research Use SEDRA as a structured data source for digital humanities research on Syriac language and literature. The two endpoints expose deterministic JSON for word and lexeme lookups, suitable for scripted batch enrichment of historical manuscripts or critical editions. Combined with a manuscript transcription pipeline, SEDRA can resolve every word to its lexeme entry overnight. Example prompt: Iterate every transcribed word in a manuscript JSON file and call `/word/{form}.json` to attach SEDRA grammatical metadata ### AI Agent for Linguistic Lookups An AI agent uses Jentic to discover SEDRA's word and lexeme operations and answer Syriac language questions in chat. The agent searches by intent, loads the operation schema from Jentic, and executes the lookup against SEDRA without parsing the bespoke ReDoc page. Because SEDRA is unauthenticated, integration via Jentic is essentially zero-config. Example prompt: Search Jentic for 'look up syriac word' and execute the returned `/word/{id}` operation with the user's input form ## Key endpoints | Method | Path | Description | | --- | --- | --- | | GET | `/word/{id}` | Look up a Syriac word by SEDRA id or Syriac text | | GET | `/lexeme/{id}` | Look up a lexeme by id | ## Key resources - **Word** — Syriac word lookup by SEDRA word id or by Syriac text in any vocalisation - **Lexeme** — Lexeme lookup by id, returning the dictionary headword and its metadata ## Why Jentic - **Setup:** SEDRA IV needs no credentials, but wiring it by hand still means composing the word and lexeme lookup URLs, choosing the accepted input form, and parsing the Syriac response yourself. Through Jentic you install once, import the SEDRA IV API from the API Directory, and your agent calls it with consistent execution handling. - **Permission scoping:** SEDRA IV is read-only public data with its word and lexeme ids in the URL path (`/word/{id}` and `/lexeme/{id}`), so scope the agent to the operations it needs, such as a word or lexeme lookup. You choose that operation set, and there are no write operations to add. - **Credential handling:** SEDRA IV is a public unauthenticated API, so there are no secrets to store. Requests still run through your own Jentic One instance for consistent observability and rate-limit awareness. - **Discovery method:** Agents search Jentic by intent such as 'look up a Syriac word', and Jentic returns the `/word/{id}` operation with its parameter schema, including the accepted SEDRA id, vocalised, or consonantal forms, so the agent calls the right endpoint without browsing the reference docs. ## Related APIs - **Wordnik API** — English language dictionary lookup as an analogue for SEDRA's Syriac lookups - **Wikimedia API** — Cross-reference Syriac words with Wiktionary entries and source manuscripts ## FAQ ### Why is there no official OpenAPI spec for SEDRA IV API? Beth Mardutho does not publish an OpenAPI specification. Jentic generates and maintains this spec so that AI agents and developers can call SEDRA IV API via structured tooling. It is validated against the live API and kept up to date. Get started with Jentic One, the self-hosted execution layer. ### What authentication does the SEDRA IV API use? The SEDRA IV API requires no authentication - it is a public read-only endpoint with CORS enabled, and responses are wildcard same-origin. There are no credentials to store in your Jentic One instance for this API; agents call it directly through Jentic with no key configuration. ### Can I look up a Syriac word by its consonantal form? Yes. GET `/word/{id}` accepts a SEDRA word id, a fully vocalised form, a partially vocalised form, or a consonantal form. Accuracy is highest for SEDRA ids, followed by fully vocalised, partially vocalised, and consonantal forms in that order. ### How do I retrieve a SEDRA word entry through Jentic? Run pip install jentic and search for 'look up syriac word'. Jentic returns the GET `/word/{id}` operation, the agent loads the schema, and executes with either the numeric id (for example 30862) or the Syriac text. The response is JSON with grammatical and lexeme references. ### What rate limits apply to the SEDRA IV API? The OpenAPI specification does not document explicit rate limits. SEDRA is a research service, so be considerate with bulk lookups - cache responses locally and avoid hammering the live host with overlapping concurrent requests. ### Is the SEDRA IV API free to use? Yes. SEDRA is provided by the Beth Mardutho project as an open research resource, with no API key or commercial tier. CORS is enabled so it can be called from the browser, and the same endpoints are available to scripted clients via Jentic. ### Can I limit what my agent is allowed to do with the SEDRA IV API? Yes. Because you run Jentic One yourself, your own rules decide which SEDRA IV operations the agent may call, and you can scope it to just the read-only lookups it needs, such as the word lookup at GET `/word/{id}` or the lexeme lookup at GET `/lexeme/{id}.` Since SEDRA IV is public data with the word and lexeme ids in the URL path and has no write operations, the operation set you choose is the full extent of what the agent can touch. You control that operation set in your own instance.