Document Parameter Consistency Review
Paste the following prompt into your AI chat to install this skill:
Please install @user_1b22b3b1/spec-param-verify according to https://skillhub.cn/install/skillhub.md.
About this skill
Problem
Engineering design documents often repeat the same parameter across reports or chapters, such as normal storage level, total storage capacity, design discharge, dam crest elevation, and installed capacity. Manual review requires a maintained parameter list, entity disambiguation, and localization of whether a conflict is intra-chapter, cross-chapter, or cross-document, which can easily miss cross-document references.
How the skill works
auto-parameter-consistency-review performs consistency checks on .docx documents (.doc can be converted or parsed in a suitable environment) without relying on parameter tables, feature sheets, or external knowledge bases. It extracts parameter name, value, and unit patterns from prose, tables, and composite dimensions; normalizes units and parameter names; and groups entries by entity + parameter name. Only parameters that appear at least twice are reported. Same-named parameters for different structures, equipment instances, schemes, or operating conditions are separated, preventing direct comparison between Gate A and Gate B or a main pipe and a branch pipe.
Key steps include:
- Parsing documents into blocks JSON, and splitting by chapter when the result exceeds roughly 100k characters.
- Building a shared multi-document context covering the project, structures, equipment, scheme status, and document discipline.
- Running a secondary conflict review that filters single-character noise and preserves cross-document constraints for global parameters.
- Using LLM review to validate grouping, add non-standard parameter expressions, and mark the minimum conflict scope.
- Generating review-result.md and review-result.xlsx with statistics, conflicts, review-required items, and evidence locations.
Boundary and notes
The skill is best suited for repeated parameters across engineering texts, especially water levels, storage, discharge, elevations, installed capacity, meteorological, seismic, electromechanical, and structural parameters. It does not review single-occurrence parameters or invent values absent from the documents. When ownership, instance, scheme, or condition is unclear, it reports the item for review instead of marking it as a definitive conflict.
Use Cases
- Compare repeated water levels and storage capacities across hydrology and hydraulic reports.
- Disambiguate same-named electromechanical parameters by unit or equipment instance.
- Locate cross-document conflicts for design discharge with chapter paths and source excerpts.
- Generate a Markdown report and Excel conflict list after reviewing multiple .docx files.
Best For
- Hydraulic design engineers checking reports for inconsistent feature water levels across documents.
- Electromechanical reviewers who need same-named ratings split by unit or pipe instance.
- Project technical leads who need conflicts, review items, and evidence before review.
- Documentation QA staff who need .docx review results consolidated into Excel.
Related Skills
Generate and edit .pptx decks with python-pptx, applying structured layouts, design rules, native charts, and visual QA to reduce template-like output.
Tencent Cloud Table Recognition V3 is an OCR skill for detecting and recognizing tables in images or PDFs, supporting various table types like linear and borderless tables, with Excel export.
The complete set of online document operation tools provided by Tencent Docs MCP, supporting creation, querying, and editing of smart docs, Excel, PPT, mind maps, and more.
Turn a topic, notes, or tables into a complete PPT plan with logic, page layouts, visual and chart guidance, and natural defense scripts.