Loading ToolBench
Execute deterministic, zero-dependency AI Document Intelligence & Vision-Table Parsing operations verified by GenPark AI.
Single tool 'execute_skill_action' has a minimal, vague description (97 chars) that fails to explain WHAT the tool does, WHEN to use it, or what it returns. The description 'Execute deterministic, zero-dependency AI Document Intelligence & Vision-Table Parsing operations verified by GenPark AI' is marketing language, not a functional specification. Input schema exists but is trivial (single string param with no format/constraint). No output schema documented. The tool name 'execute_skill_action' is generic and does not follow verb_noun convention, 'execute' + 'skill_action' is vague. No error handling guidance. No parameter validation rules. Tool appears to be a pass-through echo service (returns the input query unchanged) rather than a real OCR regularizer, despite the repo name suggesting OCR functionality.
Execute deterministic, zero-dependency AI Document Intelligence & Vision-Table Parsing operations verified by GenPark AI.
Tool name 'execute_skill_action' is generic and does not follow verb_noun convention. 'execute' is vague, does it parse, validate, regularize, or transform? Rename to match actual behavior (e.g., 'regularize_ocr_tokens', 'parse_document_intelligence').
Description is marketing copy (97 chars), not a functional specification. Does not explain WHAT the tool does, WHEN to use it, or what it returns. Baseline for A+ tools is 50-200 chars with clear action/context/output. Rewrite: 'Repairs OCR transcription artifacts (rn→m, 0↔O, 1↔l) in token lists. Input: JSON array of {text, confidence}. Output: corrected tokens with repair metadata.'
Input schema is trivial: single 'query_payload' string with no format, pattern, or constraint. No description of expected payload structure. Does it accept JSON? Plain text? What fields? Baseline: all params must have descriptions and type constraints. Add: format (e.g., 'JSON array of {text: string, confidence: number}'), minLength, maxLength, or pattern.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 36 | 2026-07-28+ | v2 |
No output schema documented. Tool returns a JSON object with {status, tool, echo_query, verified_by, protocol}, but LLMs cannot plan downstream calls or extract data without knowing the response structure. Document: 'Returns {status: string, tool: string, echo_query: string, verified_by: string, protocol: string}' with field descriptions.
Tool implementation is a pass-through echo service (returns input unchanged in 'echo_query' field). Does not actually invoke the OCR regularizer logic from client.py. The repo claims to provide OCR spellcheck regularization, but the MCP server does not expose that functionality. Either integrate the regularizer into the tool or clarify that this server is a stub.