Four tools with complete input schemas and reasonable descriptions (avg 156 chars). All parameters are typed and described. Naming follows verb_noun convention (search_*, get_*). However, descriptions lack LLM-optimized guidance on WHEN to use each tool vs. alternatives, and output schemas are not documented. No error handling guidance visible. Parameter descriptions are functional but generic, missing format constraints, ranges, and dependency hints. No tool annotations (readOnlyHint, etc.) despite all being read-only operations.

• Document output schemas for all tools. Example: 'Returns {projects: [{id, title, researchers, abstract, funding, ...}], total_count, offset}'. This enables LLMs to chain calls and extract required fields.
• Add WHEN-to-use guidance to descriptions. E.g., 'Use this to find projects by keyword, researcher name, or institution. For a specific project ID, use get_project_detail instead.'
• Add readOnlyHint: true annotation to all four tools via FastMCP's tool registration to signal safety for retry and caching.
• Specify numeric ranges in parameter descriptions: 'limit (1 - 200, default 20)', 'fiscal_year_from (1965 - 2024)', 'offset (≥0)'.
• Add error recovery hints: 'If project not found, verify the project ID format or try search_projects with keywords.' Include this in get_project_detail description.