xml-reader — Claude Code skill: parse construction XML into DataFrames
xml-reader — a skill for Claude Code, Cursor, Codex and other platforms that reads XML from P6, IFC, COBie and buildingSMART Data Dictionary. Converts elements to pandas DataFrames.
MIT 125 installs ★ 434 updated 2026-10-19
npx skills add datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --skill xml-reader -a claude-codeWhat xml-reader does
Parses P6 schedules (XER format), IFC-XML building models, COBie-XML facility data and BSDD exports. Built with Python's xml.etree.ElementTree, it extracts elements, attributes and nested nodes, then converts them into rows and columns.
For P6 files it pulls activities, resources and WBS into separate tables. For COBie it splits data by sheet (Facility, Floor, Space, Component). For IFC-XML it lists all entities with the Ifc prefix. Requires Python 3 and pandas.
What it helps with
- parse a P6 schedule into a table with tasks and dates
- extract all spaces and components from COBie-XML
- turn IFC-XML into a DataFrame for analysis
- read a buildingSMART Data Dictionary export
- pull attributes and text from nested XML elements
Example prompts for the agent
- Parse
schedule.xmlfrom P6 and list all tasks with start dates - Read
building.cobie.xmland show all spaces with their areas - Load IFC-XML from
model.ifcxmland count how manyIfcWallelements it has
How to install xml-reader
1Skills CLI (Vercel)
npx skills add datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --skill xml-reader -a claude-code2Manually: put the files in place
~/.claude/skills/xml-reader/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/xml-reader/
1Skills CLI (Vercel)
npx skills add datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --skill xml-reader -a codex2Manually: put the files in place
.agents/skills/xml-reader/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --skill xml-reader -a cursor2Manually: put the files in place
.cursor/skills/xml-reader/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction 2_DDC_Book/2.1-Data-Types-Classification/xml-reader2Manually: put the files in place
.github/skills/xml-reader/SKILL.md1Skills CLI (Vercel)
npx skills add datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --skill xml-reader -a gemini-cli2Platform CLI
gemini skills install https://github.com/datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --path 2_DDC_Book/2.1-Data-Types-Classification/xml-readerManually: put the files in place
.devin/skills/xml-reader/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --skill xml-reader -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about xml-reader
How do I install xml-reader in Claude Code?
Run npx skills add datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction --skill xml-reader -a claude-code, then open ~/.claude/skills/xml-reader/SKILL.md.
What does xml-reader do?
It reads and parses XML from construction systems — P6 schedules, IFC models, COBie data, BSDD dictionaries. Converts elements into pandas DataFrames for analysis and processing.
What does xml-reader need to run?
Python 3 and the pandas library. The skill uses the built-in xml.etree.ElementTree module for parsing.
Source text
SKILL.md · MIT
---
name: "xml-reader"
description: "Read and parse XML from construction systems - P6 schedules, BSDD exports, IFC-XML, COBie-XML. Convert to pandas DataFrames."
homepage: "https://datadrivenconstruction.io"
metadata: {"openclaw": {"emoji": "🏷️", "os": ["darwin", "linux", "win32"], "homepage": "https://datadrivenconstruction.io", "requires": {"bins": ["python3"]}}}
---
# XML Reader for Construction Data
## Overview
XML is used in construction for P6 schedules (XER), IFC-XML, COBie-XML, and buildingSMART Data Dictionary exports. This skill parses XML and converts to structured DataFrames.
## Python Implementation
```python
import xml.etree.ElementTree as ET
import pandas as pd
from typing import Dict, Any, List, Optional, Union
from dataclasses import dataclass
from pathlib import Path
import re
@dataclass
class XMLElement:
"""Parsed XML element."""
tag: str
attributes: Dict[str, str]
text: Optional[str]
children: List['XMLElement']
class ConstructionXMLReader:
"""Parse XML from construction systems."""
def __init__(self):
self.namespaces: Dict[str, str] = {}
def parse_file(self, file_path: str) -> ET.Element:
"""Parse XML file and return root element."""
tree = ET.parse(file_path)
root = tree.getroot()
# Extract namespaces
self._extract_namespaces(root)
return root
def parse_string(self, xml_string: str) -> ET.Element:
"""Parse XML from string."""
root = ET.fromstring(xml_string)
self._extract_namespaces(root)
return root
def _extract_namespaces(self, root: ET.Element):
"""Extract namespace mappings."""
# Find namespace declarations
for attr, value in root.attrib.items():
if attr.startswith('{'):
ns = attr[1:attr.index('}')]
self.namespaces[root.tag.split('}')[0][1:]] = ns
def find_elements(self, root: ET.Element,
tag: str,
namespace: str = None) -> List[ET.Element]:
"""Find all elements with given tag."""
if namespace:
tag = f"{{{namespace}}}{tag}"
return root.findall(f".//{tag}")
def element_to_dict(self, element: ET.Element,
include_children: bool = True) -> Dict[str, Any]:
"""Convert element to dictionary."""
result = {
'_tag': element.tag.split('}')[-1] if '}' in element.tag else element.tag,
'_text': element.text.strip() if element.text else None,
**element.attrib
}
if include_children:
for child in element:
child_tag = child.tag.split('}')[-1] if '}' in child.tag else child.tag
if child_tag in result:
# Multiple children with same tag - make list
if not isinstance(result[child_tag], list):
result[child_tag] = [result[child_tag]]
result[child_tag].append(self.element_to_dict(child))
else:
result[child_tag] = self.element_to_dict(child)
return result
def elements_to_dataframe(self, elements: List[ET.Element]) -> pd.DataFrame:
"""Convert list of elements to DataFrame."""
records = []
for elem in elements:
record = {'_tag': elem.tag.split('}')[-1]}
record.update(elem.attrib)
# Get direct text content
if elem.text and elem.text.strip():
record['_text'] = elem.text.strip()
# Get child values
for child in elem:
child_tag = child.tag.split('}')[-1]
if child.text and child.text.strip():
record[child_tag] = child.text.strip()
# Also get child attributes
for attr, val in child.attrib.items():
record[f"{child_tag}_{attr}"] = val
records.append(record)
return pd.DataFrame(records)
def flatten_xml(self, root: ET.Element,
target_tag: str = None) -> pd.DataFrame:
"""Flatten XML to DataFrame."""
if target_tag:
elements = self.find_elements(root, target_tag)
else:
elements = list(root)
return self.elements_to_dataframe(elements)
class P6XMLReader(ConstructionXMLReader):
"""Reader for Primavera P6 XML exports."""
def parse_activities(self, root: ET.Element) -> pd.DataFrame:
"""Parse activities from P6 XML."""
activities = self.find_elements(root, 'Activity')
return self.elements_to_dataframe(activities)
def parse_resources(self, root: ET.Element) -> pd.DataFrame:
"""Parse resources from P6 XML."""
resources = self.find_elements(root, 'Resource')
return self.elements_to_dataframe(resources)
def parse_wbs(self, root: ET.Element) -> pd.DataFrame:
"""Parse WBS from P6 XML."""
wbs = self.find_elements(root, 'WBS')
return self.elements_to_dataframe(wbs)
def parse_full_schedule(self, file_path: str) -> Dict[str, pd.DataFrame]:
"""Parse complete P6 schedule."""
root = self.parse_file(file_path)
return {
'activities': self.parse_activities(root),
'resources': self.parse_resources(root),
'wbs': self.parse_wbs(root)
}
class IFCXMLReader(ConstructionXMLReader):
"""Reader for IFC-XML files."""
def parse_entities(self, root: ET.Element) -> pd.DataFrame:
"""Parse IFC entities."""
# Find all Ifc* elements
all_entities = []
for elem in root.iter():
if elem.tag.startswith('Ifc'):
all_entities.append(elem)
return self.elements_to_dataframe(all_entities)
def get_entity_types(self, root: ET.Element) -> Dict[str, int]:
"""Count entity types."""
counts = {}
for elem in root.iter():
tag = elem.tag
if tag.startswith('Ifc'):
counts[tag] = counts.get(tag, 0) + 1
return counts
class COBieXMLReader(ConstructionXMLReader):
"""Reader for COBie XML files."""
COBIE_SHEETS = ['Facility', 'Floor', 'Space', 'Zone', 'Type',
'Component', 'System', 'Assembly', 'Connection',
'Spare', 'Resource', 'Job', 'Document', 'Attribute']
def parse_cobie(self, file_path: str) -> Dict[str, pd.DataFrame]:
"""Parse all COBie sheets."""
root = self.parse_file(file_path)
result = {}
for sheet in self.COBIE_SHEETS:
elements = self.find_elements(root, sheet)
if elements:
result[sheet] = self.elements_to_dataframe(elements)
return result
class BSDDXMLReader(ConstructionXMLReader):
"""Reader for buildingSMART Data Dictionary exports."""
def parse_classifications(self, root: ET.Element) -> pd.DataFrame:
"""Parse classification items."""
items = self.find_elements(root, 'Classification')
return self.elements_to_dataframe(items)
def parse_properties(self, root: ET.Element) -> pd.DataFrame:
"""Parse property definitions."""
props = self.find_elements(root, 'Property')
return self.elements_to_dataframe(props)
```
## Quick Start
```python
reader = ConstructionXMLReader()
# Parse XML file
root = reader.parse_file("schedule.xml")
# Find specific elements
activities = reader.find_elements(root, "Activity")
print(f"Found {len(activities)} activities")
# Convert to DataFrame
df = reader.elements_to_dataframe(activities)
```
## Common Use Cases
### 1. P6 Schedule Import
```python
p6_reader = P6XMLReader()
schedule = p6_reader.parse_full_schedule("p6_export.xml")
activities = schedule['activities']
print(f"Activities: {len(activities)}")
```
### 2. COBie Data
```python
cobie_reader = COBieXMLReader()
cobie_data = cobie_reader.parse_cobie("facility_cobie.xml")
components = cobie_data.get('Component', pd.DataFrame())
```
### 3. IFC-XML Analysis
```python
ifc_reader = IFCXMLReader()
root = ifc_reader.parse_file("model.ifcxml")
# Count entity types
types = ifc_reader.get_entity_types(root)
for entity_type, count in sorted(types.items(), key=lambda x: -x[1])[:10]:
print(f"{entity_type}: {count}")
```
## Resources
- **DDC Book**: Chapter 2.1 - Semi-structured Data
- **IFC-XML**: buildingSMART specification
© datadrivenconstruction · MIT · datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction