Loading ToolBench

Reading the current corpus…

bible-ko-mcp MCP server · C · ToolBench

← Browse

bible-ko-mcp

github.com/oksure/bible-ko-mcp

MCP server (and Claude Skill) for the Korean Bible (bskorea.or.kr)

2026-07-28+STDIO6 tools

69/100

Definition Quality79

Protocol Readiness43

SupportabilityNot measured-

Spec posture

Inferred effective spec 2026-07-28+·current baseline 2026-07-28

Static source inference · medium confidence

Current-spec patterns detected

Summary

The server defines 6 tools with explicit JSON schemas and descriptions. All tools are READ_ONLY, which is appropriate for a Bible lookup service. Naming follows verb_noun conventions (get_, search_, list_, compare_) which is solid. However, descriptions are terse (67-161 chars), parameter descriptions lack depth, and output schemas are not formally documented. Error handling is basic (returns text errors but doesn't provide recovery guidance or categorization). No parameter validation error messages shown. The 'search-bible' tool explicitly disclaims it only searches the first 10 chapters, a limitation that should be reflected in parameter descriptions for clarity. Overall: serviceable but underspecified for optimal LLM interaction.

Tools (6)

compare-translationsread onlysource verified82/100

Compare a verse across different Korean translations

get-chapterread onlysource verified83/100

Get all verses from a specific chapter of the Korean Bible

get-versesread onlysource verified83/100

Get specific verse(s) from a chapter

health-checkread onlysource verified70/100

Check the health status of the Bible Korean MCP server and API connectivity

list-booksread onlysource verified77/100

List all available books in the Bible

search-bibleread onlysource verified78/100

Search for verses containing specific keywords. Searches the first 10 chapters of each book — not a full-Bible search.

Issues (5)

HIGHtool-description ↗

Parameter descriptions are minimal or missing depth. Book parameter accepts 'English or Korean' names and codes but provides no guidance on format, case sensitivity, or fallback behavior if exact match fails. Verse and chapter numbers lack range context (min is specified but max is not).

get-chapterget-versescompare-translations

HIGHresponse-shaper ↗

Output schemas are not documented. Code returns formatted text (markdown) but LLMs cannot infer structure. No specification of what fields a Chapter or Verse object contains, or what the search-bible result structure is. This forces LLMs to parse freeform text rather than operating on structured data.

get-chapterget-versessearch-biblecompare-translations

MEDIUMrecovery-guide ↗

Error handling returns plain text messages without recovery guidance. E.g. 'Book not found' suggests calling list-books, but the error message does not include this suggestion. No error categorization (retryable vs user-fixable vs fatal).

get-chapter

Recommendations

Add explicit output schemas to every tool's definition. Example: 'Returns a Chapter object: {book: string, bookKorean: string, chapter: number, versionName: string, verses: [{number: number, text: string}]}'
Expand parameter descriptions with constraints and examples. For 'book': 'Book name (English or Korean) or code (e.g., Genesis, 창세기, gen). Case-insensitive. If no exact match, will attempt fuzzy matching.'
Add max verse/chapter limits to numeric parameters. E.g. 'chapter: Chapter number (1 - 66 for OT, 1 - 4 for NT depending on book)' or document this in the response if dynamic.
Enhance error messages with actionable recovery steps. Change 'Book not found' to 'Book not found. Use list-books to see available books, or search-bible to find verses by keyword.'
Document the search-bible limitation more prominently. Add to description: 'Note: Searches only the first 10 chapters of each book. For a specific chapter/verse, use get-chapter or get-verses instead.'
Add a 'versions' parameter to get-chapter and get-verses to allow multi-version comparison inline, reducing need for separate compare-translations calls.
Consider renaming 'health-check' to 'check-api-health' or 'diagnose-connection' to clarify it is a monitoring/diagnostic tool, not part of the core Bible lookup flow.
Add type specifications for list-books response. E.g. 'Returns an array of {code: string, english: string, korean: string, testament: "OT"|"NT"}'. Currently the response structure is opaque.
Validate book codes and provide a helpful message if a user passes an invalid code, rather than a silent 404.

Score history

Overall score trend

↑ 25 points across a rubric change (v1 → v2)

69/100

Scored	Grade	Overall	Spec posture	Rubric
2026-09-22	C	69	2026-07-28+	v2
2026-03-09	F	44	-	v1