Loading ToolBench
Read-only scripture-study engine exposing Greek/Hebrew biblical text, morphology, lexicon, and concordance data over MCP (Model Context Protocol)
Orthotomeo demonstrates solid tool design with 11 well-named, verb-prefixed tools (get_verse, concord_lemma, lemmatize, etc.) serving a specialized biblical scholarship domain. All tools have descriptions (avg ~120 chars) and explicit JSON schemas with typed parameters. However, output schemas are not documented in the visible code, parameter descriptions lack constraint details (ranges, enums, formats), and error handling guidance is absent. The tools follow single-responsibility principle well and chain naturally (lemmatize → lexicon_lookup). No security issues detected (read-only operations, no secrets). Naming is domain-appropriate but could be more discoverable for LLMs unfamiliar with biblical terminology.
Get morphological attestation (case, tense, voice, mood, etc.) for a word at a specific verse location
Render a set of citations as a pastable reference block in Markdown format
Find all occurrences of a lemma, disambiguated Strong's number, or inflected word form across the corpus
Find all verses containing an ordered sequence of lemmas within a specified word distance
Count total occurrences of a lemma, dStrong, or surface form across the corpus, with breakdown by book
Fetch a range of verses' text from specified editions
Fetch one verse's text from specified editions (KJV, ASV, WEB, BSB, Brenton)
Output schemas not documented. Tools return structured data but LLMs cannot see what fields to expect, forcing them to infer structure from examples or fail on unexpected formats.
Parameter descriptions lack constraint details. 'editions' accepts 5 specific values (KJV, ASV, WEB, BSB, Brenton) but description does not enumerate them; 'window' in concord_phrase lacks min/max bounds; 'word' in attest lacks 1-based indexing clarification.
No error handling guidance. Tools do not document what happens on invalid book codes, out-of-range verses, or missing corpus data. LLMs cannot self-correct or know whether to retry.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 56 | 2026-07-28+ | v2 |
Build an interlinear display (word + gloss + morphology) for a verse or word range
Get lemma and Strong's number for each word in a verse
Look up a lexicon entry by disambiguated Strong's number to get gloss, definition, transliteration, and language
Resolve a scripture reference to its canonical form and check if it exists in the database
Ambiguous parameter 'by' in concord_lemma and count. Description says 'optional match column override' but does not explain what auto-detect does or when to use each value (lemma vs dstrong vs surface).
Domain-specific terminology ('lemma', 'dStrong', 'TAGNT', 'TAHOT') in descriptions assumes LLM familiarity with biblical scholarship. No glossary or context provided for agents unfamiliar with the domain.