Critical gaps across naming, descriptions, and schemas. Tool names lack action verbs (e.g., 'edgar_peer_comparison' vs 'analyze_peer_comparison'). Descriptions exist but are domain-jargon-heavy and lack actionable guidance for LLM selection. No parameter descriptions visible for tools accepting parameters. Error handling and recovery guidance absent. This server is below the 50th percentile for community MCP tools.

• Rename tools with action verbs: 'edgar_peer_comparison' → 'analyze_peer_comparison', 'edgar_trend_break_analysis' → 'detect_trend_breaks', 'profile_dataset' → 'profile_columns', 'build_panel' → 'build_financial_panel', 'run_pipeline' → 'run_edgar_analysis_pipeline'.
• Add complete input schemas for all 12 tools. For tools with no parameters, use an empty object schema with description 'No parameters required.' For parameterized tools, define type, description, and constraints for each field (e.g., 'ticker': {type: 'string', description: 'Stock ticker symbol (e.g., AAPL, MSFT)', pattern: '^[A-Z]{1,5}$'}).
• Rewrite descriptions to be LLM-optimized (50-200 chars). Example: 'edgar_peer_comparison' → 'Compare a company\'s financial metrics against industry peers using SEC EDGAR data. Returns peer percentiles and outlier flags.' Include WHEN to use and WHAT to expect.
• Add parameter descriptions for all 9 parameterized tools. Example for 'ticker': 'Stock ticker symbol (e.g., AAPL). Required to resolve company CIK and fetch SEC filings.' Example for 'refresh': 'Force fresh data fetch from SEC API instead of using cached results. Default: false.'
• Add error handling and recovery guidance to tool descriptions. Example: 'If ticker not found, try search_companies() with partial name. If SEC API times out, retry with refresh=false to use cache.' Include retryability classification (retryable vs user-fixable vs fatal).