Critical failures across all dimensions. No input schemas visible in source code for any of the 6 tools. Tool descriptions are present but extremely brief (10-30 chars), far below the 50-200 char baseline for LLM-optimized descriptions. No parameter documentation visible. No output schemas documented. Tools appear to be inferred from index.ts references rather than explicitly registered with full definitions. This is a framework integration (OpenCode plugin), not a standalone MCP server with proper tool definitions.

• Expand each tool description to 80-150 characters. Include: what the tool does, when to use it (vs similar tools), what it returns, and any prerequisites. Example: 'Run project test suite and return pass/fail counts, coverage %, and failed test names. Use before committing code. Returns structured test results with per-suite breakdown.'
• Define explicit input schemas for all tools in JSON Schema format. Include type, description, enum constraints, and min/max for each parameter. Example: { 'type': 'object', 'properties': { 'test_pattern': { 'type': 'string', 'description': 'Glob pattern for test files (e.g., **/*.test.ts). Defaults to all tests.' } } }
• Document output schemas. Specify what fields each tool returns, their types, and what they mean. Example for run-tests: { 'type': 'object', 'properties': { 'passed': { 'type': 'integer' }, 'failed': { 'type': 'integer' }, 'coverage_percent': { 'type': 'number' }, 'failed_tests': { 'type': 'array', 'items': { 'type': 'string' } } } }
• Add error handling guidance to descriptions. State what can go wrong and what the LLM should do. Example: 'If tests fail, returns error details. If no tests found, returns empty result, check test_pattern parameter.'
• Verify tool definitions are explicitly registered in the MCP server code with full schemas, not inferred from file references. Provide the actual tool registration code showing names, descriptions, input schemas, and output schemas.