Three tools with Turkish/English descriptions and basic schemas. ada_workflow has reasonable parameter documentation (komut, gorev, lang with types and descriptions). ada_komutlar and ada_kurallar are simpler but lack depth. No input validation guidance, no error recovery hints, no output schema documentation. Descriptions are present but generic (10-50 chars for ada_komutlar/ada_kurallar). No tool annotations (readOnlyHint, destructiveHint, idempotentHint). Parameter descriptions exist but lack constraints (e.g., lang enum not enforced in schema). No pagination, no chaining IDs, no structured output specs. All tools are READ_ONLY but this is not formally declared in tool metadata.

• Add readOnlyHint=true to all three tools in the @mcp.tool() decorator to formally declare they are safe to call without side effects.
• Document output schema for ada_workflow: specify that it returns a string containing workflow instructions, behavior rules, and reference format. For ada_komutlar, document that it returns a markdown-formatted command map with sections and command lists.
• Convert 'lang' parameter to an enum constraint in JSON Schema: {"type": "string", "enum": ["tr", "en"], "description": "Output language..."}. This prevents invalid values and is machine-parseable.
• Expand descriptions for ada_komutlar and ada_kurallar to 100-150 chars each, explaining when to call them (e.g., 'Call this first to discover available ADA commands before invoking ada_workflow') and what structure they return.
• Add error recovery guidance to ada_workflow description: 'If komut is not recognized, call ada_komutlar() to see valid commands. If gorev context is unclear, provide customer/project details.'