schedule-extractor

Name: schedule-extractor
Author: dleerdefi

24stars5forks

How to install schedule-extractor?

dleerdefi/claude-code-construction/schedule-extractor

$npx -y skills add dleerdefi/claude-code-construction --skill schedule-extractor

Installs into the current project.

Summary

Extract tabular schedule data from construction drawings — door, window, finish, fixture, panel schedules — and output to Excel. Triggers: 'door schedule', 'extract schedule', 'schedule to Excel', 'panel schedule'.

Files · 1

View on GitHub

SKILL.md

1	# Schedule Extractor
2
3	Extracts tabular schedule data embedded in drawing sheets or spec pages and outputs structured Excel files. Schedules are typically one element among many on a sheet — or the entire sheet may be a schedule.
4
5	## Workflow
6
7	```
8	Extraction Progress:
9	- [ ] Step 1: Discover schedule locations
10	- [ ] Step 2: Isolate (crop) the schedule region
11	- [ ] Step 3: Extract structured data (pdfplumber + vision)
12	- [ ] Step 4: Validate and clean data
13	- [ ] Step 5: Output to Excel
14	- [ ] Step 6: Write graph entry
15	```
16
17	### Step 1: Discover Schedule Locations
18
19	Use a multi-source discovery approach, checking all available sources:
20
21	#### Source A — Sheet titles in sheet index (primary, most reliable)
22
23	Search `.construction/index/sheet_index.yaml` (or the sheet index you've built) for sheets with "SCHEDULE" in the title. Many important schedules occupy an entire sheet — the sheet title tells you exactly what it is:
24	- "DOOR SCHEDULE" → entire sheet is a door schedule
25	- "FINISH SCHEDULE" or "ROOM FINISH SCHEDULE" → entire sheet is finishes
26	- "WINDOW SCHEDULE" → entire sheet is windows
27	- "PANEL SCHEDULE" → electrical panel schedule
28	- "FIXTURE SCHEDULE" → plumbing fixtures
29
30	For dedicated schedule sheets, the entire page is the extraction target — no need to crop.
31
32	#### Source B — AgentCM database query (supplementary)
33
34	If `.construction/database.yaml` exists, read `query_command` from it, then query known schedules:
35	```bash
36	# Read query_command and project_id from .construction/database.yaml
37	{query_command} -c "SELECT id, schedule_type, title, sheet_id, bounding_region FROM schedules WHERE project_id = '{PROJECT_ID}'"
38	```
39	This returns all schedules already detected (including stubs from Group Review with bounding regions). For embedded schedules on non-schedule sheets, check the `bounding_region` column.
40
41	Note: Schedule bounding region detection is still being refined — treat these as hints, not definitive boundaries. Always verify with vision.
42
43	#### Source C — Discipline-based heuristics
44
45	Schedules appear on specific sheet types:
46	- Door schedule → typically on A-0.XX or A-8.XX sheets, or a dedicated sheet
47	- Window schedule → same sheets as door schedule, or separate
48	- Room finish schedule → A-0.XX or interior sheets
49	- Panel schedule → E-X.XX electrical sheets
50	- Fixture schedule → P-X.XX plumbing sheets
51	- Equipment schedule → M-X.XX mechanical sheets
52
53	#### Source D — Vision scan (fallback)
54
55	If no index or graph is available:
56	```bash
57	${CLAUDE_SKILL_DIR}/../../bin/construction-python ${CLAUDE_SKILL_DIR}/../../scripts/pdf/rasterize_page.py {pdf_path} {page} --dpi 150 --output full_sheet.png
58	```
59
60	Use vision on the full sheet image: "Identify any tabular schedules on this drawing sheet. Report the approximate bounding box coordinates (top-left x,y and bottom-right x,y) as percentages of the image dimensions, the schedule type, and the column headers visible."
61
62	### Step 2: Isolate the Schedule Region
63
64	For dedicated schedule sheets (entire page is a schedule): Skip cropping — use the full page.
65
66	For embedded schedules (schedule is one element on a larger sheet):
67
68	Crop the identified region with padding:
69	```bash
70	${CLAUDE_SKILL_DIR}/../../bin/construction-python ${CLAUDE_SKILL_DIR}/../../scripts/pdf/crop_region.py full_sheet.png \
71	--box {x1},{y1},{x2},{y2} \
72	--padding 20 \
73	--output schedule_crop.png
74	```
75
76	Re-rasterize at higher DPI (300) for the cropped region to improve text clarity:
77	```bash
78	${CLAUDE_SKILL_DIR}/../../bin/construction-python ${CLAUDE_SKILL_DIR}/../../scripts/pdf/rasterize_page.py {pdf_path} {page} \
79	--dpi 300 \
80	--crop {x1},{y1},{x2},{y2} \
81	--output schedule_hires.png
82	```
83
84	### Step 3: Extract Structured Data
85
86	Use a try → validate → fallback approach:
87
88	#### Method A — pdfplumber table extraction (try first)
89
90	Use pdfplumber's `extract_tables()` method on the target page. If multiple tables are found, select the largest one (most rows with the most columns — schedules are wide). The first row of the selected table contains headers; subsequent rows are data. If no tables are found or results look garbled, fall back to vision (Method B).
91
92	#### Evaluate Method A — Quality Gate
93
94	Check these criteria before proceeding:
95
96	1. Row count: Did pdfplumber extract at least 10 data rows? Most schedules have 20-200+ entries.
97	2. Column consistency: Do ≥80% of rows have the same number of columns? Inconsistent columns = parsing error.
98	3. Non-empty cells: Are >50% of cells non-empty? Mostly-empty extraction = parsing failed.
99	4. First column validity: Does the first column contain recognizable IDs (door numbers, ro

People who liked this also viewed

6 picks

Type

schedule-extractor

24stars5forks

View on GitHub

How to install schedule-extractor?

dleerdefi/claude-code-construction/schedule-extractor

$npx -y skills add dleerdefi/claude-code-construction --skill schedule-extractor

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Schedule Extractor
2
3	Extracts tabular schedule data embedded in drawing sheets or spec pages and outputs structured Excel files. Schedules are typically one element among many on a sheet — or the entire sheet may be a schedule.
4
5	## Workflow
6
7	```
8	Extraction Progress:
9	- [ ] Step 1: Discover schedule locations
10	- [ ] Step 2: Isolate (crop) the schedule region
11	- [ ] Step 3: Extract structured data (pdfplumber + vision)
12	- [ ] Step 4: Validate and clean data
13	- [ ] Step 5: Output to Excel
14	- [ ] Step 6: Write graph entry
15	```
16
17	### Step 1: Discover Schedule Locations
18
19	Use a multi-source discovery approach, checking all available sources:
20
21	#### Source A — Sheet titles in sheet index (primary, most reliable)
22
23	Search `.construction/index/sheet_index.yaml` (or the sheet index you've built) for sheets with "SCHEDULE" in the title. Many important schedules occupy an entire sheet — the sheet title tells you exactly what it is:
24	- "DOOR SCHEDULE" → entire sheet is a door schedule
25	- "FINISH SCHEDULE" or "ROOM FINISH SCHEDULE" → entire sheet is finishes
26	- "WINDOW SCHEDULE" → entire sheet is windows
27	- "PANEL SCHEDULE" → electrical panel schedule
28	- "FIXTURE SCHEDULE" → plumbing fixtures
29
30	For dedicated schedule sheets, the entire page is the extraction target — no need to crop.
31
32	#### Source B — AgentCM database query (supplementary)
33
34	If `.construction/database.yaml` exists, read `query_command` from it, then query known schedules:
35	```bash
36	# Read query_command and project_id from .construction/database.yaml
37	{query_command} -c "SELECT id, schedule_type, title, sheet_id, bounding_region FROM schedules WHERE project_id = '{PROJECT_ID}'"
38	```
39	This returns all schedules already detected (including stubs from Group Review with bounding regions). For embedded schedules on non-schedule sheets, check the `bounding_region` column.
40
41	Note: Schedule bounding region detection is still being refined — treat these as hints, not definitive boundaries. Always verify with vision.
42
43	#### Source C — Discipline-based heuristics
44
45	Schedules appear on specific sheet types:
46	- Door schedule → typically on A-0.XX or A-8.XX sheets, or a dedicated sheet
47	- Window schedule → same sheets as door schedule, or separate
48	- Room finish schedule → A-0.XX or interior sheets
49	- Panel schedule → E-X.XX electrical sheets
50	- Fixture schedule → P-X.XX plumbing sheets
51	- Equipment schedule → M-X.XX mechanical sheets
52
53	#### Source D — Vision scan (fallback)
54
55	If no index or graph is available:
56	```bash
57	${CLAUDE_SKILL_DIR}/../../bin/construction-python ${CLAUDE_SKILL_DIR}/../../scripts/pdf/rasterize_page.py {pdf_path} {page} --dpi 150 --output full_sheet.png
58	```
59
60	Use vision on the full sheet image: "Identify any tabular schedules on this drawing sheet. Report the approximate bounding box coordinates (top-left x,y and bottom-right x,y) as percentages of the image dimensions, the schedule type, and the column headers visible."
61
62	### Step 2: Isolate the Schedule Region
63
64	For dedicated schedule sheets (entire page is a schedule): Skip cropping — use the full page.
65
66	For embedded schedules (schedule is one element on a larger sheet):
67
68	Crop the identified region with padding:
69	```bash
70	${CLAUDE_SKILL_DIR}/../../bin/construction-python ${CLAUDE_SKILL_DIR}/../../scripts/pdf/crop_region.py full_sheet.png \
71	--box {x1},{y1},{x2},{y2} \
72	--padding 20 \
73	--output schedule_crop.png
74	```
75
76	Re-rasterize at higher DPI (300) for the cropped region to improve text clarity:
77	```bash
78	${CLAUDE_SKILL_DIR}/../../bin/construction-python ${CLAUDE_SKILL_DIR}/../../scripts/pdf/rasterize_page.py {pdf_path} {page} \
79	--dpi 300 \
80	--crop {x1},{y1},{x2},{y2} \
81	--output schedule_hires.png
82	```
83
84	### Step 3: Extract Structured Data
85
86	Use a try → validate → fallback approach:
87
88	#### Method A — pdfplumber table extraction (try first)
89
90	Use pdfplumber's `extract_tables()` method on the target page. If multiple tables are found, select the largest one (most rows with the most columns — schedules are wide). The first row of the selected table contains headers; subsequent rows are data. If no tables are found or results look garbled, fall back to vision (Method B).
91
92	#### Evaluate Method A — Quality Gate
93
94	Check these criteria before proceeding:
95
96	1. Row count: Did pdfplumber extract at least 10 data rows? Most schedules have 20-200+ entries.
97	2. Column consistency: Do ≥80% of rows have the same number of columns? Inconsistent columns = parsing error.
98	3. Non-empty cells: Are >50% of cells non-empty? Mostly-empty extraction = parsing failed.
99	4. First column validity: Does the first column contain recognizable IDs (door numbers, ro

People who liked this also viewed

6 picks

Type