Loading ToolBench
MCP server for converting .docx, .html, and .md files to .hwpx format using pypandoc-hwpx
Three tools with consistent structure and reasonable descriptions (120-150 chars each), but significant gaps in schema completeness and parameter documentation. All tools follow verb_noun naming (docx_to_hwpx, html_to_hwpx, md_to_hwpx). Input schemas present but lack type constraints (no enums, ranges, or format specs). Parameters have descriptions but are sparse (20-40 chars). No output schema documented. Error handling is minimal, no recovery guidance or categorization. Tools are well-composed (single responsibility) but lack the rigor expected for production use.
Convert a .docx file to a .hwpx file. Args: input_path: Local file path or URL to the source .docx file. output_path: Path where the resulting .hwpx file will be written. Required for local input; optional when input_path is a URL and Azure Blob Storage is configured. reference_hwpx: Optional path to a reference .hwpx file for styles. Returns: The absolute path of the generated .hwpx file, or a download URL when the source was a URL and Azure Blob Storage is configured.
Convert an HTML file to a .hwpx file. Args: input_path: Local file path or URL to the source HTML file. output_path: Path where the resulting .hwpx file will be written. Required for local input; optional when input_path is a URL and Azure Blob Storage is configured. reference_hwpx: Optional path to a reference .hwpx file for styles. Returns: The absolute path of the generated .hwpx file, or a download URL when the source was a URL and Azure Blob Storage is configured.
Convert a Markdown file to a .hwpx file. Args: input_path: Local file path or URL to the source Markdown (.md) file. output_path: Path where the resulting .hwpx file will be written. Required for local input; optional when input_path is a URL and Azure Blob Storage is configured. reference_hwpx: Optional path to a reference .hwpx file for styles. Returns: The absolute path of the generated .hwpx file, or a download URL when the source was a URL and Azure Blob Storage is configured.
Output schema not documented. Tools return a string (file path or URL) but LLMs cannot infer downstream chaining requirements or field structure.
Parameters lack type constraints. 'input_path' and 'output_path' accept free-form strings with no validation hints (file extension, URL format, path length). No enums, patterns, or min/max bounds.
Error handling absent. No guidance on recovery (e.g., 'If conversion fails, check file format' or 'Retry with a different reference_hwpx'). No error categorization (retryable vs. fatal).
Parameter descriptions are terse (20-40 chars). 'Local file path or URL to the source .docx file' lacks guidance on format, size limits, or when to use URL vs. local path.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 58 | <=2025-11-25 | v2 |
No idempotency or confirmation pattern for destructive operations. Tools write files (WRITE risk) but lack dry-run or confirmation step to prevent accidental overwrites.