Streaming availability across 200+ services: titles, sources, releases.
This server has critical gaps across naming, descriptions, and schemas. Tool names lack action verbs (title_search is acceptable, but title_detail, title_sources, title_seasons are vague nouns). Descriptions are extremely terse (1-3 words: 'Search titles.', 'Title detail.', 'Seasons.'), well below the 10-character minimum and far below the 194-char baseline for A+ tools. Parameter descriptions are missing entirely for most tools (e.g., title_search has no descriptions for search_field or types; releases has 8 params with zero descriptions). Output schemas are completely undocumented, LLMs cannot plan downstream calls or extract needed fields. Error handling is absent; the code throws generic errors like 'Watchmode: 401' with no recovery guidance. No tool annotations (readOnlyHint, etc.). The server is a thin API wrapper with minimal agent-optimization.
Genres.
Languages.
Title catalog.
Networks.
Regions.
Recent/upcoming releases.
Streaming sources/services.
Descriptions are critically short (1 - 3 words). 'Title detail.', 'Seasons.', 'Genres.' provide zero context for LLM tool selection. Baseline is 194 chars; these are 5 - 15 chars.
Parameter descriptions are completely missing. title_search has search_field and types with no descriptions; releases has 8 params with zero descriptions. LLMs cannot infer what 'search_field' means or what values 'types' accepts.
Output schemas are completely undocumented. No tool declares what fields it returns, what types they are, or how to chain results to downstream tools. LLMs cannot plan multi-step workflows or extract needed IDs.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 35 | 2026-07-28+ | v2 |
Title detail.
Search titles.
Seasons.
Streaming sources.
Vague noun-based naming (title_detail, title_sources, title_seasons) lacks action verbs. LLMs infer intent from names; 'title_detail' could mean get, describe, or fetch. Should be get_title_detail or describe_title. Baseline: 90% of A+ tools start with action verbs.
No error recovery guidance. Code throws 'Watchmode: 401' or 'Watchmode: 500' with no actionable next steps. LLMs cannot self-correct or retry intelligently. Should return 'Invalid API key. Verify PLATFORM_WATCHMODE_KEY is set.' with recovery hints.