Model Context Protocol server for accessing anime data through AniDB HTTP API
This server provides 7 tools across anime search and details with generally adequate naming and description coverage, but significant gaps in schema visibility, parameter documentation, and output specification. Tool names follow verb_noun convention (search, get, trending) which is good. However, the evaluation is severely constrained by inability to verify actual input/output schemas in the source code provided. The tools appear to be dynamically registered via a provider pattern (anime_search_*, anime_details_*, etc.), making actual schema definitions difficult to confirm. While descriptions are present and reasonably detailed (150-250 chars), they lack specificity around output structure, error handling, and parameter constraints. No error handling guidance is evident. The server lacks output schema documentation entirely, agents cannot predict what fields will be returned. Parameters like 'limit' and 'season' lack min/max constraints in visible descriptions.
Get detailed information about a specific anime from AniDB. Retrieve comprehensive anime data from AniDB including synopsis, ratings, episode count, air dates, creators, related anime information, and enhanced data such as episodes, external resources, characters, tags, and recommendations. Data Source: AniDB (https://anidb.net)
Search for anime by title using AniDB local database. Search the AniDB database for anime matching the provided query string. Uses a local database for fast response times and reduced API dependency. Returns basic information about matching anime including title, type, and year. Data Source: AniDB (https://anidb.net)
Get detailed information about a specific anime from the provider. Retrieve comprehensive anime data from the provider including synopsis, ratings, episode count, air dates, creators, and related anime information.
Get anime recommendations based on a specific anime from the provider. Retrieve recommended anime related to a given title from the provider.
Search for anime by title using the provider. Search the provider database for anime matching the provided query string. Returns basic information about matching anime including title, type, and year.
Output schemas completely undocumented. No tool specifies what fields are returned, data types, or structure. LLMs cannot plan downstream operations or extract required fields for tool chaining.
Dynamic tool registration via provider pattern (anime_search_*, anime_details_*, etc.) means actual schema definitions are not visible in provided source.
Parameter constraints missing. 'limit' parameters lack minimum/maximum (e.g., is 20 valid? 1000?). 'season' parameter accepts free-form string instead of enum constraint (spring|summer|fall|winter). 'query' minimum length stated as 2 chars but not enforced or described for all search tools.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | F | 43 | <=2025-11-25 | v2 |
| 2026-03-09 | C | 61 | - | v1 |
Get seasonal anime for a specific season and year from the provider. Retrieve anime releases for a specified season from the provider.
Get trending anime from the provider. Retrieve currently trending anime based on provider metrics.
No error handling guidance. Tools lack documentation on what errors can occur, whether they are retryable, or what the agent should do if a search returns zero results, a timeout occurs, or an invalid anime_id is provided.
Descriptions do not specify output structure or chaining capability. For example, 'anidb_search' description does not state whether it returns anime_id (needed for anidb_details), title, year, or type. Without knowing output fields, agents cannot chain tools.
Result limits not enforced or described. 'anime_search' says 'max: 20' but 'anime_seasonal' says 'max varies by provider', vague and inconsistent. No guidance on what happens if an agent requests 1000 results.
Generic parameter descriptions. For example, 'aid' in anidb_details is described as 'must be a positive integer' but does not explain where to obtain it, what range is valid (1 - 999999?), or what happens if an invalid ID is provided.