Loading ToolBench
Single tool 'parse_torrent' has a reasonable description (127 chars, within 10-1024 range) but critical schema deficiencies. Input parameters are documented in description text but lack formal JSON Schema type definitions and constraints. No output schema is documented, LLMs cannot predict return structure. Parameter descriptions are present but lack format constraints (e.g., base64 encoding, URL validation rules, magnet URI format). No error handling guidance. Tool does one thing (parse torrent metadata) but schema incompleteness and missing output documentation severely limit agent usability.
Parse torrent file or magnet link and extract metadata including name, info hash, file list, size, trackers, and health information (seeds/peers)
No formal JSON Schema for input parameters. Parameters (data, file, url) are described in prose but lack type, format, and constraint declarations. LLMs cannot validate inputs or understand mutually-exclusive parameter logic.
Output schema completely undocumented. Response includes magnetUri, name, info_hash, size, seeds, peers, trackers, files, etc., but no schema definition. LLMs cannot plan downstream operations or extract fields reliably.
Mutually-exclusive parameters (data, file, url) lack explicit documentation of mutual exclusivity. LLMs may pass multiple parameters simultaneously, causing ambiguous behavior.
No error handling guidance. Tool can fail (invalid torrent, network timeout, SSRF rejection, file too large) but returns no recovery hints. LLMs receive raw errors with no actionable next steps.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 41 | 2026-07-28+ | v2 |
Tool name 'parse_torrent' is acceptable but generic. Does not clearly signal that it also fetches from URLs, handles magnet links, and returns health data (seeds/peers). Name could be 'extract_torrent_metadata' for clarity.