Loading ToolBench

Reading the current corpus…

egw-writings-mcp.todelete MCP server · D · ToolBench

egw-writings-mcp.todelete

github.com/GospelSounders/egw-writings-mcp.todelete

MCP server providing access to Ellen G. White (EGW) writings with local database support, API integration, and PDF generation capabilities

2026-07-28+STDIO13 tools

55/100

Definition Quality73

Protocol Readiness50

Supportability28

Spec posture

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

Static source inference · medium confidence

Current-spec patterns detected

Summary

Server provides 13 read-only tools with complete JSON schemas and basic descriptions. All tools have proper input schemas with types and descriptions, meeting baseline requirements. However, descriptions are generic and lack LLM-optimized guidance on when/why to use each tool. No output schemas documented. Error handling is minimal. Tool naming is clear and verb-based (search_, get_, list_, browse_), following conventions well. Parameter descriptions exist but lack constraints (enums, ranges, formats). No tool annotations (readOnlyHint, idempotentHint) despite all being read-only. Missing composition guidance for multi-step workflows.

Tools (13)

browse_by_referenceread onlysource verified77/100

Find content by EGW reference (e.g., "AA 15.1", "GC 123")

get_book_contentread onlysource verified75/100

Get chapter content from a book

get_book_detailsread onlysource verified73/100

Get detailed information about a specific book

get_contextread onlysource verified73/100

Get surrounding paragraphs for better context

get_database_statsread onlysource verified68/100

Get statistics about the local database

get_local_bookread onlysource verified72/100

Get book information from local database

get_local_contentread onlysource verified73/100

Get paragraphs from a book in the local database

get_suggestionsread only

Issues (5)

HIGHtool ↗

No output schemas documented. LLMs cannot plan downstream calls or extract required fields without knowing response structure. All 13 tools lack documented return types.

search_writingsget_book_detailsget_book_contentlist_languageslist_booksget_suggestionssearch_localget_local_bookget_local_contentlist_local_booksget_database_statsbrowse_by_referenceget_context

HIGHtool-description ↗

Descriptions lack LLM-optimized guidance. Generic descriptions like 'Get detailed information about a specific book' do not explain WHEN to use this tool vs alternatives (e.g., get_book_details vs get_book_content vs get_local_book). Average description length ~60 chars; baseline is 194 chars.

get_book_detailslist_languageslist_booksget_suggestionsget_local_booklist_local_booksget_database_stats

Recommendations

Document output schemas for all 13 tools. Specify field names, types, and whether results are paginated. Example: search_writings returns {results: [{id, title, author, excerpt, language}], total_count, has_more}.
Expand descriptions to 100-200 chars with WHEN/WHY guidance. Example: 'Search Ellen G. White writings via live API. Use this for current/complete results; use search_local for offline/faster searches. Returns up to 20 results; use offset for pagination.'
Add tool annotations: mark all tools with readOnlyHint=true and idempotentHint=true to signal safety and retry semantics to agents.
Add parameter constraints: language as enum of supported codes (en, es, fr, etc.); limit with min=1, max=100; offset with min=0; reference with pattern regex for EGW format (e.g., '^[A-Z]{2} \d+(\.\d+)?$').
Consolidate overlapping tools or add explicit disambiguation. Either merge API/local variants into single tools with a 'source' parameter, or add descriptions explaining: 'Use search_writings for live API access; use search_local for offline database. search_local is faster but may be outdated.'

Score history

Overall score trend

First recorded score · v2 rubric

55/100

Scored	Grade	Overall	Spec posture	Rubric
2026-09-23	D	55	2026-07-28+	v2