Critical failure across all dimensions. Tool definitions are inferred from module names only, no explicit schema, parameter types, or descriptions visible in source. The six tools (discovery_tool_1/2, invocation_tool_1/2, development_tool_1/2) have generic placeholder names that violate verb_noun convention and provide zero actionable guidance to LLMs. Descriptions are trivial (15-20 chars, below 20-char floor). No input schemas, output schemas, or parameter documentation found. Tool registration functions exist (register_discovery_tools, etc.) but actual tool definitions are not visible in provided source. Per HARD SCORING RULE: tools with no visible schema → schema score 0; descriptions under 20 chars → description score 0-20; inferred-only tools → cap at 50. This server cannot be reliably used by agents.

• Rename tools with action verbs: discovery_tool_1 → search_agents, discovery_tool_2 → list_available_agents, invocation_tool_1 → invoke_agent, invocation_tool_2 → get_agent_result, development_tool_1 → create_agent, development_tool_2 → deploy_agent. Each name must clearly state what the tool does.
• Add 50-200 character descriptions to each tool explaining WHAT it does, WHEN to use it, and what it returns. Example: 'Search for agents in the Catallaxy marketplace by name, category, or rating. Returns agent ID, name, description, price, and rating. Use this to discover available agents before invoking.'
• Document input schemas with JSON Schema: define all parameters with type (string, number, boolean, array, object), required fields, enums for constrained values, and min/max for numeric ranges. Example: {"agent_name": {"type": "string", "description": "Agent name or partial name to search", "minLength": 1, "maxLength": 100}}
• Document output schemas showing all returned fields, their types, and meanings. Include pagination fields (total_count, next_cursor) if applicable. Example: {"agents": [{"agent_id": "string", "name": "string", "price_nanoton": "integer", "rating": "number"}], "total_count": "integer"}
• Add error handling guidance: categorize errors as retryable vs user-fixable, include the invalid value and constraint violated, and suggest recovery steps. Example: 'Agent not found. Try search_agents() with a partial name or browse available agents with list_available_agents()'