document-export

Name: document-export
Author: chacosoldier

29stars6forks

How to install document-export?

chacosoldier/compabob/document-export

$npx -y skills add chacosoldier/compabob --skill document-export

Installs into the current project.

Summary

Create a PDF, Excel (.xlsx), or Word (.docx) file from content. Use for "/document-export", "make a PDF", "export to Excel", "put this in a spreadsheet", "save this as a Word doc", or when the user needs a shareable office document rather than a note or a raw HTML page.

Files · 1

View on GitHub

SKILL.md

1	# Document Export
2
3	Turn content into a real office file: a PDF, an Excel workbook, or a
4	Word document. The skill ships three helper scripts that do the file writing
5	deterministically; your job is to assemble the content and call the right one.
6
7	These scripts use the Python standard library only, so there is nothing to
8	install for Excel and Word. PDF needs a browser or a PDF tool on the machine
9	(the script finds it, or tells the user what to install).
10
11	Save every output to `reports/` as `YYYY-MM-DD-<slug>.<ext>`, then tell the user
12	the path.
13
14	## PDF
15
16	PDF is produced by rendering an HTML file. Two cases:
17
18	- Rich, visual page (architecture, dashboard, comparison): use the
19	`visual-explainer` skill to build the HTML, then convert it.
20	- Plain document (a letter, a report, a write-up): wrap the content in this
21	minimal template and save it as an `.html` file:
22
23	```html
24	<!DOCTYPE html>
25	<html><head><meta charset="utf-8"><title>TITLE</title>
26	<style>
27	@page { margin: 2cm; }
28	body { font: 11pt/1.55 -apple-system, "Segoe UI", Roboto, sans-serif;
29	color: #1a1a1a; max-width: 46em; margin: 0 auto; }
30	h1, h2, h3 { line-height: 1.25; }
31	h1 { font-size: 1.8em; } h2 { font-size: 1.35em; margin-top: 1.6em; }
32	table { border-collapse: collapse; width: 100%; }
33	th, td { border: 1px solid #ccc; padding: 6px 10px; text-align: left; }
34	code { background: #f4f4f4; padding: 1px 4px; border-radius: 3px; }
35	</style></head><body>
36	<!-- CONTENT -->
37	</body></html>
38	```
39
40	Then convert:
41
42	```bash
43	python3 .claude/skills/document-export/to_pdf.py <input.html> reports/<YYYY-MM-DD>-<slug>.pdf
44	```
45
46	It tries headless Chrome first, then weasyprint, then wkhtmltopdf. If it reports
47	that no engine is found, relay its install hint to the user; do not try to work
48	around it.
49
50	## Excel (.xlsx)
51
52	Build a JSON spec, write it to a temporary file, and run the writer.
53
54	Spec shape:
55
56	```json
57	{
58	"sheets": [
59	{
60	"name": "Results",
61	"header": true,
62	"rows": [["Name", "Score"], ["Alice", 90], ["Bob", 85]],
63	"widths": [24, 10]
64	}
65	]
66	}
67	```
68
69	- `header: true` bolds the first row. `widths` (column widths in characters) is
70	optional. Multiple sheets are allowed.
71	- Cell values may be strings, numbers, booleans, or `null` (an empty cell).
72	Keep numbers as numbers in the JSON so Excel treats them as numeric.
73
74	Run it:
75
76	```bash
77	python3 .claude/skills/document-export/to_xlsx.py /tmp/export-spec.json reports/<YYYY-MM-DD>-<slug>.xlsx
78	```
79
80	You can also pipe the spec on stdin by passing `-` instead of a file path.
81
82	## Word (.docx)
83
84	Build a JSON spec of blocks, then run the writer.
85
86	Spec shape:
87
88	```json
89	{
90	"title": "Optional document title",
91	"blocks": [
92	{"type": "heading", "level": 2, "text": "A section"},
93	{"type": "paragraph", "text": "A plain paragraph."},
94	{"type": "paragraph", "runs": [
95	{"text": "mixed ", "bold": true},
96	{"text": "formatting", "italic": true}
97	]},
98	{"type": "bullets", "items": ["first point", "second point"]},
99	{"type": "numbered", "items": ["step one", "step two"]},
100	{"type": "table", "header": true, "rows": [["Col A", "Col B"], ["1", "2"]]}
101	]
102	}
103	```
104
105	- `heading` levels are 1 to 3. A `paragraph` carries plain `text` (with optional
106	`bold` / `italic` for the whole paragraph) or a list of `runs` for mixed
107	formatting. `table` rows are lists of cell strings; `header: true` bolds the
108	first row.
109
110	Run it:
111
112	```bash
113	python3 .claude/skills/document-export/to_docx.py /tmp/export-spec.json reports/<YYYY-MM-DD>-<slug>.docx
114	```
115
116	## After exporting
117
118	- Delete any temporary spec file you created.
119	- Give the user the exact path of the file.
120	- If the user wants to share the file, that is an outward action: hand it over,
121	do not send it anywhere yourself.
122
123	## Scope
124
125	These writers cover the structure that carries content: values, headings,
126	formatting, lists, simple tables, column widths. They do not do Excel formulas,
127	charts, or merged cells, or Word images, footnotes, or page headers. If a
128	request needs those, say so and either extend the script or, if the user has the
129	`openpyxl` or `python-docx` library available, write a one-off script with it
130	instead.

People who liked this also viewed

6 picks

Type

document-export

29stars6forks

View on GitHub

How to install document-export?

chacosoldier/compabob/document-export

$npx -y skills add chacosoldier/compabob --skill document-export

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Document Export
2
3	Turn content into a real office file: a PDF, an Excel workbook, or a
4	Word document. The skill ships three helper scripts that do the file writing
5	deterministically; your job is to assemble the content and call the right one.
6
7	These scripts use the Python standard library only, so there is nothing to
8	install for Excel and Word. PDF needs a browser or a PDF tool on the machine
9	(the script finds it, or tells the user what to install).
10
11	Save every output to `reports/` as `YYYY-MM-DD-<slug>.<ext>`, then tell the user
12	the path.
13
14	## PDF
15
16	PDF is produced by rendering an HTML file. Two cases:
17
18	- Rich, visual page (architecture, dashboard, comparison): use the
19	`visual-explainer` skill to build the HTML, then convert it.
20	- Plain document (a letter, a report, a write-up): wrap the content in this
21	minimal template and save it as an `.html` file:
22
23	```html
24	<!DOCTYPE html>
25	<html><head><meta charset="utf-8"><title>TITLE</title>
26	<style>
27	@page { margin: 2cm; }
28	body { font: 11pt/1.55 -apple-system, "Segoe UI", Roboto, sans-serif;
29	color: #1a1a1a; max-width: 46em; margin: 0 auto; }
30	h1, h2, h3 { line-height: 1.25; }
31	h1 { font-size: 1.8em; } h2 { font-size: 1.35em; margin-top: 1.6em; }
32	table { border-collapse: collapse; width: 100%; }
33	th, td { border: 1px solid #ccc; padding: 6px 10px; text-align: left; }
34	code { background: #f4f4f4; padding: 1px 4px; border-radius: 3px; }
35	</style></head><body>
36	<!-- CONTENT -->
37	</body></html>
38	```
39
40	Then convert:
41
42	```bash
43	python3 .claude/skills/document-export/to_pdf.py <input.html> reports/<YYYY-MM-DD>-<slug>.pdf
44	```
45
46	It tries headless Chrome first, then weasyprint, then wkhtmltopdf. If it reports
47	that no engine is found, relay its install hint to the user; do not try to work
48	around it.
49
50	## Excel (.xlsx)
51
52	Build a JSON spec, write it to a temporary file, and run the writer.
53
54	Spec shape:
55
56	```json
57	{
58	"sheets": [
59	{
60	"name": "Results",
61	"header": true,
62	"rows": [["Name", "Score"], ["Alice", 90], ["Bob", 85]],
63	"widths": [24, 10]
64	}
65	]
66	}
67	```
68
69	- `header: true` bolds the first row. `widths` (column widths in characters) is
70	optional. Multiple sheets are allowed.
71	- Cell values may be strings, numbers, booleans, or `null` (an empty cell).
72	Keep numbers as numbers in the JSON so Excel treats them as numeric.
73
74	Run it:
75
76	```bash
77	python3 .claude/skills/document-export/to_xlsx.py /tmp/export-spec.json reports/<YYYY-MM-DD>-<slug>.xlsx
78	```
79
80	You can also pipe the spec on stdin by passing `-` instead of a file path.
81
82	## Word (.docx)
83
84	Build a JSON spec of blocks, then run the writer.
85
86	Spec shape:
87
88	```json
89	{
90	"title": "Optional document title",
91	"blocks": [
92	{"type": "heading", "level": 2, "text": "A section"},
93	{"type": "paragraph", "text": "A plain paragraph."},
94	{"type": "paragraph", "runs": [
95	{"text": "mixed ", "bold": true},
96	{"text": "formatting", "italic": true}
97	]},
98	{"type": "bullets", "items": ["first point", "second point"]},
99	{"type": "numbered", "items": ["step one", "step two"]},
100	{"type": "table", "header": true, "rows": [["Col A", "Col B"], ["1", "2"]]}
101	]
102	}
103	```
104
105	- `heading` levels are 1 to 3. A `paragraph` carries plain `text` (with optional
106	`bold` / `italic` for the whole paragraph) or a list of `runs` for mixed
107	formatting. `table` rows are lists of cell strings; `header: true` bolds the
108	first row.
109
110	Run it:
111
112	```bash
113	python3 .claude/skills/document-export/to_docx.py /tmp/export-spec.json reports/<YYYY-MM-DD>-<slug>.docx
114	```
115
116	## After exporting
117
118	- Delete any temporary spec file you created.
119	- Give the user the exact path of the file.
120	- If the user wants to share the file, that is an outward action: hand it over,
121	do not send it anywhere yourself.
122
123	## Scope
124
125	These writers cover the structure that carries content: values, headings,
126	formatting, lists, simple tables, column widths. They do not do Excel formulas,
127	charts, or merged cells, or Word images, footnotes, or page headers. If a
128	request needs those, say so and either extend the script or, if the user has the
129	`openpyxl` or `python-docx` library available, write a one-off script with it
130	instead.

People who liked this also viewed

6 picks

Type