Critical failures across naming, descriptions, and schemas. 31 tools with minimal documentation. Most tool names lack action verbs (e.g., 'build-index', 'fetch-docs', 'cache-status' instead of 'build_search_index', 'fetch_documentation', 'check_cache_status'). Descriptions are extremely brief (10-50 chars), providing no context for LLM selection. Input schemas visible for only ~8 tools; most lack parameter type definitions. No output schemas documented. No error handling guidance. Tool definitions appear inferred from shell scripts rather than explicitly registered with proper MCP schemas.

• Rename all tools to verb_noun format: 'build-index' → 'build_search_index', 'fetch-docs' → 'fetch_documentation', 'cache-status' → 'check_cache_status'. This enables LLMs to infer intent from the name alone.
• Expand all descriptions to 50-200 characters. Include WHAT the tool does, WHEN to use it, and what it returns. Example: 'semantic-search' should be 'Search documentation using semantic similarity. Returns matching docs with relevance scores. Use when keyword search is insufficient.'
• Add explicit JSON Schema input definitions for all 31 tools. Every parameter must have a 'type' field (string, integer, boolean, array, object) and a 'description' field explaining its purpose and constraints.
• Document output schemas for all tools. Specify the structure of returned data (fields, types, required vs optional). This enables LLMs to chain tools and extract the right data for downstream calls.
• Add error handling guidance to tool descriptions. For each tool, document what can go wrong and what the LLM should do next. Example: 'If search returns no results, try broader keywords or use list_all_docs to discover available topics.'