spec-splitter

Name: spec-splitter
Author: dleerdefi

24stars5forks

How to install spec-splitter?

dleerdefi/claude-code-construction/spec-splitter

$npx -y skills add dleerdefi/claude-code-construction --skill spec-splitter

Installs into the current project.

Summary

Split a bound project manual PDF into individual spec section PDFs and extract searchable text. Triggers: 'split specs', 'break up the project manual', 'separate spec sections', 'extract spec text'. Prerequisite for /submittal-log-generator.

Files · 1

View on GitHub

SKILL.md

1	# Spec Splitter
2
3	Two functions for specification processing:
4
5	1. Split: Break a bound project manual PDF into individual spec section PDFs — navigable files the project team can use directly
6	2. Extract: Pull searchable text from each section PDF into persistent `.txt` files — enables downstream skills (submittal-log-generator, spec-parser) to work from text without re-extracting from PDFs
7
8	## Pipeline Position
9	Run after `/project-setup` identifies bound spec manuals. Produces split PDFs, `spec_index.yaml`, and extracted text consumed by `/submittal-log-generator` and `/code-researcher`.
10
11	Either function can run independently. For example, specs may already be split but text has not yet been extracted.
12
13	## Workflow
14
15	```
16	Spec Split Progress:
17	- [ ] Step 1: Check current state (split? text extracted?)
18	- [ ] Step 2: Discover Specifications directory
19	- [ ] Step 3: Find ALL spec PDFs (bound manuals)
20	- [ ] Step 4-5: Split PDF into individual section files
21	- [ ] Step 6: Write spec index
22	- [ ] Step 7: Extract text from all sections
23	- [ ] Step 8: Repair degraded/poor text quality
24	- [ ] Step 9: Write graph entry (AgentCM only)
25	```
26
27	### Step 1: Check Current State
28
29	Check what already exists:
30
31	Split PDFs present?
32	- Look for individual spec section PDFs with CSI section numbers in filenames (e.g., `03 30 00 - Cast-in-Place Concrete.pdf`)
33	- Check for `spec_index.yaml`
34	- If found, report count and skip to Step 7 (text extraction)
35
36	Text already extracted?
37	- Check `.construction/spec_text/manifest.json`
38	- If manifest exists and covers all sections, report and skip Step 7
39
40	### Step 2: Discover Specifications Directory
41
42	Determine where split spec PDFs should go. Search for an existing Specifications directory (case-insensitive):
43	1. `02 - Specifications/` (numbered project folder convention)
44	2. `Specifications/`
45	3. Any folder with "specification" in the name
46
47	Output directory resolution:
48	- If Specifications directory found → output to `{specs_dir}/Specification Sections/`
49	- If not found → output to `Specification Sections/` in project root
50
51	### Step 3: Find ALL Spec PDFs
52
53	Search the project directory for ALL PDFs that are specifications. Many projects have multiple spec PDFs:
54	- Multi-volume: Volume 1.pdf, Volume 2.pdf (split by CSI division range)
55	- Single bound manual: one large PDF with all sections
56	- Attachment-based: Attachment-E-Specs.pdf (government projects)
57
58	Search in:
59	- The Specifications directory discovered in Step 2
60	- The project root (some projects have no folder structure)
61	- Look for PDFs > 1MB with keywords: "spec", "manual", "volume", "attachment" + spec-related terms
62
63	Process EACH PDF found. All split sections go to the same output directory.
64
65	If the user specifies a single file (`/spec-splitter path/to/specific-volume.pdf`), process only that file.
66
67	### Steps 4-6: Split and Index
68
69	Run the split script with the resolved output directory:
70
71	```bash
72	${CLAUDE_SKILL_DIR}/../../bin/construction-python ${CLAUDE_SKILL_DIR}/scripts/split_spec_manual.py \
73	"{project_manual.pdf}" \
74	--output-dir "{resolved_spec_sections_dir}"
75	```
76
77	The script:
78	1. Scans ALL pages for `SECTION XX XX XX` headers to find exact page boundaries — this is the primary method and does NOT depend on a Table of Contents
79	2. Scans all pages for Table of Contents entries to enrich section titles (optional, best-effort)
80	3. For sections without ToC titles, extracts titles directly from the section header page
81	4. Splits into individual PDFs named `{section_number} - {SECTION TITLE}.pdf`
82	5. Writes `spec_index.yaml` with section metadata
83
84	ToC edge cases handled:
85	- ToC located deep in the document (e.g., page 60+): Common when front matter (transmittals, addenda) precedes the project manual. The script scans all pages, not just the first few.
86	- No ToC at all: Section boundaries are found by scanning every page for `SECTION` headers. Titles are extracted directly from each section's title page. The split still succeeds — titles may be slightly less polished than ToC-enriched versions.
87
88	### Step 7: Extract Text
89
90	After splitting (or if specs are already split), extract searchable text from every section:
91
92	```bash
93	${CLAUDE_SKILL_DIR}/../../bin/construction-python ${CLAUDE_SKILL_DIR}/scripts/extract_spec_text.py \
94	--specs-dir "{resolved_spec_sections_dir}" \
95	--output-dir ".construction/spec_text"
96	```
97
98	The script:
99	- Extracts text from each section PDF via pdfplumber
100	- Assesses extraction quality (GOOD / DEGRADED / POOR)
101	- Writes one `.txt` file per section to `.construction/spec_text/`
102	- Writes `manifest.json` with quality metadata per sec

People who liked this also viewed

6 picks

Type

spec-splitter

24stars5forks

View on GitHub

How to install spec-splitter?

dleerdefi/claude-code-construction/spec-splitter

$npx -y skills add dleerdefi/claude-code-construction --skill spec-splitter

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Spec Splitter
2
3	Two functions for specification processing:
4
5	1. Split: Break a bound project manual PDF into individual spec section PDFs — navigable files the project team can use directly
6	2. Extract: Pull searchable text from each section PDF into persistent `.txt` files — enables downstream skills (submittal-log-generator, spec-parser) to work from text without re-extracting from PDFs
7
8	## Pipeline Position
9	Run after `/project-setup` identifies bound spec manuals. Produces split PDFs, `spec_index.yaml`, and extracted text consumed by `/submittal-log-generator` and `/code-researcher`.
10
11	Either function can run independently. For example, specs may already be split but text has not yet been extracted.
12
13	## Workflow
14
15	```
16	Spec Split Progress:
17	- [ ] Step 1: Check current state (split? text extracted?)
18	- [ ] Step 2: Discover Specifications directory
19	- [ ] Step 3: Find ALL spec PDFs (bound manuals)
20	- [ ] Step 4-5: Split PDF into individual section files
21	- [ ] Step 6: Write spec index
22	- [ ] Step 7: Extract text from all sections
23	- [ ] Step 8: Repair degraded/poor text quality
24	- [ ] Step 9: Write graph entry (AgentCM only)
25	```
26
27	### Step 1: Check Current State
28
29	Check what already exists:
30
31	Split PDFs present?
32	- Look for individual spec section PDFs with CSI section numbers in filenames (e.g., `03 30 00 - Cast-in-Place Concrete.pdf`)
33	- Check for `spec_index.yaml`
34	- If found, report count and skip to Step 7 (text extraction)
35
36	Text already extracted?
37	- Check `.construction/spec_text/manifest.json`
38	- If manifest exists and covers all sections, report and skip Step 7
39
40	### Step 2: Discover Specifications Directory
41
42	Determine where split spec PDFs should go. Search for an existing Specifications directory (case-insensitive):
43	1. `02 - Specifications/` (numbered project folder convention)
44	2. `Specifications/`
45	3. Any folder with "specification" in the name
46
47	Output directory resolution:
48	- If Specifications directory found → output to `{specs_dir}/Specification Sections/`
49	- If not found → output to `Specification Sections/` in project root
50
51	### Step 3: Find ALL Spec PDFs
52
53	Search the project directory for ALL PDFs that are specifications. Many projects have multiple spec PDFs:
54	- Multi-volume: Volume 1.pdf, Volume 2.pdf (split by CSI division range)
55	- Single bound manual: one large PDF with all sections
56	- Attachment-based: Attachment-E-Specs.pdf (government projects)
57
58	Search in:
59	- The Specifications directory discovered in Step 2
60	- The project root (some projects have no folder structure)
61	- Look for PDFs > 1MB with keywords: "spec", "manual", "volume", "attachment" + spec-related terms
62
63	Process EACH PDF found. All split sections go to the same output directory.
64
65	If the user specifies a single file (`/spec-splitter path/to/specific-volume.pdf`), process only that file.
66
67	### Steps 4-6: Split and Index
68
69	Run the split script with the resolved output directory:
70
71	```bash
72	${CLAUDE_SKILL_DIR}/../../bin/construction-python ${CLAUDE_SKILL_DIR}/scripts/split_spec_manual.py \
73	"{project_manual.pdf}" \
74	--output-dir "{resolved_spec_sections_dir}"
75	```
76
77	The script:
78	1. Scans ALL pages for `SECTION XX XX XX` headers to find exact page boundaries — this is the primary method and does NOT depend on a Table of Contents
79	2. Scans all pages for Table of Contents entries to enrich section titles (optional, best-effort)
80	3. For sections without ToC titles, extracts titles directly from the section header page
81	4. Splits into individual PDFs named `{section_number} - {SECTION TITLE}.pdf`
82	5. Writes `spec_index.yaml` with section metadata
83
84	ToC edge cases handled:
85	- ToC located deep in the document (e.g., page 60+): Common when front matter (transmittals, addenda) precedes the project manual. The script scans all pages, not just the first few.
86	- No ToC at all: Section boundaries are found by scanning every page for `SECTION` headers. Titles are extracted directly from each section's title page. The split still succeeds — titles may be slightly less polished than ToC-enriched versions.
87
88	### Step 7: Extract Text
89
90	After splitting (or if specs are already split), extract searchable text from every section:
91
92	```bash
93	${CLAUDE_SKILL_DIR}/../../bin/construction-python ${CLAUDE_SKILL_DIR}/scripts/extract_spec_text.py \
94	--specs-dir "{resolved_spec_sections_dir}" \
95	--output-dir ".construction/spec_text"
96	```
97
98	The script:
99	- Extracts text from each section PDF via pdfplumber
100	- Assesses extraction quality (GOOD / DEGRADED / POOR)
101	- Writes one `.txt` file per section to `.construction/spec_text/`
102	- Writes `manifest.json` with quality metadata per sec

People who liked this also viewed

6 picks

Type