Strong schema definitions with consistent naming (verb_noun pattern: get_, search_, find_). All 11 tools have descriptions and input schemas with typed parameters. Descriptions are concise (50-150 chars) and action-oriented. However, output schemas are not documented in the source code, LLMs cannot see what fields to expect from responses. Parameter descriptions are technical but lack guidance on when to use each tool or how they chain together. No error handling guidance visible. Tool composition is clean (single responsibility), but descriptions could better explain dependencies (e.g., search_scenes returns IDs; get_scene retrieves full records).

• Document output schemas for all tools. For search_* tools, specify: returns array of {id, name, ...}. For get_* tools, specify: returns {id, name, description, ...}. Include field types and which fields are always present vs optional.
• Add error handling guidance to each tool description. E.g., 'If a catalogue is unavailable, results omit it. If no matches found, returns empty array. If hash is invalid, returns error with format requirements.'
• Clarify tool composition in descriptions. E.g., 'search_scenes returns scene IDs; call get_scene to retrieve full records. find_by_fingerprint returns matching scenes directly (no second call needed).'
• Enhance parameter descriptions with format rules. E.g., 'hash: hexadecimal string, 16 chars for OSHASH, 16 chars for PHASH, case-insensitive.' 'duration: integer, seconds, required for validation.'
• Document pagination behavior. E.g., 'Returns up to limit results per catalogue. page is 1-indexed. No total_count returned; use limit to detect if more results exist. Results from different catalogues are not merged.'

Inferred effective spec: 2025-06-18+.

• Relies on Logging (deprecated) - log to stderr or use OpenTelemetry