Two simple tools with basic schemas and minimal descriptions. Both tools have input schemas with proper types, but descriptions are terse (under 100 chars) and lack context on when/why to use them. No parameter descriptions provided, LLMs cannot infer what 'text' or 'value' mean without additional guidance. No output schema documented. Error handling is present but generic ('toml failed: ...'). Tools follow verb_noun naming (to_json, to_toml) but descriptions don't explain prerequisites, return format, or error recovery paths.

• Expand tool descriptions to 100-200 chars. Example: 'to_json: Parse TOML text and return a JSON-compatible object. Use this to convert TOML config files to JSON. Returns an object with parsed key-value pairs. Throws error if TOML is malformed.'
• Add parameter descriptions. For 'text': 'TOML-formatted string (UTF-8). Must be valid TOML 1.0 syntax.' For 'value': 'JSON object (top-level table). Arrays and primitives are rejected.'
• Document output schema in descriptions. E.g., 'to_json returns {"value": <parsed object>}. Dates/times are ISO 8601 strings.'
• Improve error messages with recovery hints. E.g., 'TOML parse error at line 5: invalid syntax. Check for missing quotes or brackets.' instead of generic 'toml failed: ...'
• Add constraint documentation for to_toml: 'value must be a JSON object (not array or primitive). Arrays-of-objects become [[table]] sections in TOML.'