Cross-discipline design-contradiction checker for construction drawing sets. Standalone tool broken out from Iron_Bid; a pipeline stage may later fold back into Iron_Bid. Pipeline: PDF->images -> per-sheet assertion extraction -> deterministic clustering by location -> per-cluster reasoning -> report. Includes CLI (cli/run_check.py) and web UI (backend/main.py). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
1 line
1.1 KiB
JSON
1 line
1.1 KiB
JSON
{"conflicts": [{"conflict_id": "RTU_Tagging_Ambiguity", "category": "tag_or_reference_inconsistency", "severity": "medium", "confidence": "high", "location": "Roof Area at Grid B-C / 4-5", "disciplines": ["Mechanical"], "sheets": ["M02.02", "M01.01"], "description": "The Mechanical Roof Plan on M02.02 shows two distinct new rooftop units, but both are identically tagged as 'RTU-1'. The Rooftop Unit Schedule on M01.01 defines the specifications for a unit type 'RTU-1'. While the units may be identical models, assigning the same tag to two separate physical objects creates ambiguity for contractors, installers, and commissioning agents. Each piece of equipment requires a unique identifier for clear reference.", "evidence": [{"discipline": "Mechanical", "sheet": "M02.02", "source_text": "Image on Mechanical Roof Plan showing two units, both labeled with the tag 'RTU-1'.", "asserted_value": "Both units are RTU-1"}, {"discipline": "Mechanical", "sheet": "M01.01", "source_text": "ROOFTOP UNIT SCHEDULE: MARK RTU-1", "asserted_value": "Defines a single unit type RTU-1"}], "recommended_resolution": "coordinate drawings"}]} |