brand-docx

Name: brand-docx
Author: ferdinandobons

24installs26forks

How to install brand-docx?

ferdinandobons/brand-docs/brand-docx

$npx -y skills add ferdinandobons/brand-docs --skill brand-docx

Installs into the current project.

Summary

Brand-aware Word engine. Use to (1) EXTRACT a company's brand from a Word template into a reusable "Brand Profile", (2) COMPREHEND the template with the model (optional), (3) VERIFY it, (4) GENERATE new on-brand .docx documents FROM a saved profile. Trigger on "extract our brand"

Files · 1

View on GitHub

SKILL.md

1	# brand-docx
2
3	Use this skill when the user wants a reusable Word brand kit or wants to create
4	a new on-brand `.docx` from a company template and variable content.
5
6	This is an AI-agent skill for Codex and Claude Code. The user should not need to
7	write JSON or run shell commands. The agent converts the user's content into an
8	IntermediateDocument, invokes the internal engine, verifies the output, and
9	returns the generated `.docx`.
10
11	## The seven verbs: three deterministic + four model-assisted
12
13	Every brand skill (`brand-docx`, `brand-pptx`, `brand-xlsx`) implements the same
14	contract. The deterministic core is extract / verify / generate; on top of it
15	sit the optional learning verbs **comprehend / learn / propose-overrides /
16	refine**, each fail-closed (the engine validates every proposal and authors every
17	value).
18
19	\| Verb \| Input \| Output \|
20	\|---\|---\|---\|
21	\| extract \| a company `.docx` template \| a reusable Brand Profile \|
22	\| comprehend (optional, model-driven) \| a saved profile + a model-authored `comprehension.json` \| the profile with a validated, cached `comprehension` block \|
23	\| verify \| a saved Brand Profile \| QA findings + a verdict \|
24	\| generate \| content (an IntermediateDocument) + a profile \| a new on-brand `.docx` \|
25	\| learn (deterministic distillation) \| the profile's cross-run generation history \| recurring QA findings distilled into shell-frozen overrides, advisory until `--accept` \|
26	\| propose-overrides (model-driven) \| the recurring remainder `learn` could not bind + a model-authored proposal \| shell-backed corrections through the same fail-closed sink, advisory until `--accept` \|
27	\| refine \| end-of-generation user feedback (text or a screenshot) as a `refinement.json` delta \| the existing comprehension overlaid for FUTURE generations, advisory until `--accept` \|
28
29	`comprehend` is optional: `generate` works on the deterministic profile alone.
30	When a current comprehension is present, `generate` additionally reconciles the
31	template's preserved cover/index structures with the new content. See
32	[reference/comprehension.md](reference/comprehension.md) for the full step.
33
34	## Hard Rules
35
36	- Treat `python scripts/cli.py ...` as an internal engine command, not the user-facing workflow.
37	- `scripts/cli.py` is a LAUNCHER that locates the engine root by itself: it works from this skill folder AND from the repo/plugin root (set `BRAND_DOCS_ROOT` to override). Never guess deeper paths like `scripts/brandkit/...`.
38	- Run the dependency preflight before starting extract / comprehend / verify / generate, and report missing or unusable dependencies before proceeding.
39	- Extract opens the source template read-only and saves `brand-kit/<name>/template/shell.docx` byte-for-byte.
40	- Generate opens the saved shell and resolves every semantic block through `profile.json`.
41	- Do not put style names, colors, fonts, or brand identifiers in an IntermediateDocument.
42	- If the user did not provide a template or enough content, ask for the missing input.
43	- Return the generated file path plus a QA summary.
44	- Consult `profile.json.artifact_catalog` before generation when the user asks to mimic a specific piece of the template.
45
46	## Preflight (always first)
47
48	Before doing any work, run:
49
50	```bash
51	python scripts/cli.py doctor
52	```
53
54	Use its output to decide the run mode:
55
56	- If a required Python dependency is missing, install/repair it before extraction
57	or generation; the core engine is not ready.
58	- If only visual renderers are missing or unusable (`soffice` plus `pdftoppm` or
59	optional PyMuPDF/`fitz`), the
60	core L0 workflow can still run, but a full visual audit cannot be claimed.
61	Tell the user what is missing, include the install/repair hint printed by
62	`doctor`, and either proceed with degraded QA or install the renderer first.
63	- If optional OCR (`tesseract`) is missing, the visual audit can still run, but
64	rendered residual-text proof is incomplete. Report that limitation when
65	judging stale placeholders or field caches.
66	- For `--qa deep` or `--qa strict`, prefer repairing/installing renderers before
67	generation. If the environment cannot run them, `deep` generates a degraded
68	manifest and `strict` fails with a visual proof blocker.
69
70	## Agent Workflow
71
72	1. Run the dependency preflight above and report any degraded capability.
73	2. Determine the brand name and locate the user-provided `.docx` template.
74	3. If no matc

People who liked this also viewed

6 picks

Type

brand-docx

24installs26forks

View on GitHub

How to install brand-docx?

ferdinandobons/brand-docs/brand-docx

$npx -y skills add ferdinandobons/brand-docs --skill brand-docx

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# brand-docx
2
3	Use this skill when the user wants a reusable Word brand kit or wants to create
4	a new on-brand `.docx` from a company template and variable content.
5
6	This is an AI-agent skill for Codex and Claude Code. The user should not need to
7	write JSON or run shell commands. The agent converts the user's content into an
8	IntermediateDocument, invokes the internal engine, verifies the output, and
9	returns the generated `.docx`.
10
11	## The seven verbs: three deterministic + four model-assisted
12
13	Every brand skill (`brand-docx`, `brand-pptx`, `brand-xlsx`) implements the same
14	contract. The deterministic core is extract / verify / generate; on top of it
15	sit the optional learning verbs **comprehend / learn / propose-overrides /
16	refine**, each fail-closed (the engine validates every proposal and authors every
17	value).
18
19	\| Verb \| Input \| Output \|
20	\|---\|---\|---\|
21	\| extract \| a company `.docx` template \| a reusable Brand Profile \|
22	\| comprehend (optional, model-driven) \| a saved profile + a model-authored `comprehension.json` \| the profile with a validated, cached `comprehension` block \|
23	\| verify \| a saved Brand Profile \| QA findings + a verdict \|
24	\| generate \| content (an IntermediateDocument) + a profile \| a new on-brand `.docx` \|
25	\| learn (deterministic distillation) \| the profile's cross-run generation history \| recurring QA findings distilled into shell-frozen overrides, advisory until `--accept` \|
26	\| propose-overrides (model-driven) \| the recurring remainder `learn` could not bind + a model-authored proposal \| shell-backed corrections through the same fail-closed sink, advisory until `--accept` \|
27	\| refine \| end-of-generation user feedback (text or a screenshot) as a `refinement.json` delta \| the existing comprehension overlaid for FUTURE generations, advisory until `--accept` \|
28
29	`comprehend` is optional: `generate` works on the deterministic profile alone.
30	When a current comprehension is present, `generate` additionally reconciles the
31	template's preserved cover/index structures with the new content. See
32	[reference/comprehension.md](reference/comprehension.md) for the full step.
33
34	## Hard Rules
35
36	- Treat `python scripts/cli.py ...` as an internal engine command, not the user-facing workflow.
37	- `scripts/cli.py` is a LAUNCHER that locates the engine root by itself: it works from this skill folder AND from the repo/plugin root (set `BRAND_DOCS_ROOT` to override). Never guess deeper paths like `scripts/brandkit/...`.
38	- Run the dependency preflight before starting extract / comprehend / verify / generate, and report missing or unusable dependencies before proceeding.
39	- Extract opens the source template read-only and saves `brand-kit/<name>/template/shell.docx` byte-for-byte.
40	- Generate opens the saved shell and resolves every semantic block through `profile.json`.
41	- Do not put style names, colors, fonts, or brand identifiers in an IntermediateDocument.
42	- If the user did not provide a template or enough content, ask for the missing input.
43	- Return the generated file path plus a QA summary.
44	- Consult `profile.json.artifact_catalog` before generation when the user asks to mimic a specific piece of the template.
45
46	## Preflight (always first)
47
48	Before doing any work, run:
49
50	```bash
51	python scripts/cli.py doctor
52	```
53
54	Use its output to decide the run mode:
55
56	- If a required Python dependency is missing, install/repair it before extraction
57	or generation; the core engine is not ready.
58	- If only visual renderers are missing or unusable (`soffice` plus `pdftoppm` or
59	optional PyMuPDF/`fitz`), the
60	core L0 workflow can still run, but a full visual audit cannot be claimed.
61	Tell the user what is missing, include the install/repair hint printed by
62	`doctor`, and either proceed with degraded QA or install the renderer first.
63	- If optional OCR (`tesseract`) is missing, the visual audit can still run, but
64	rendered residual-text proof is incomplete. Report that limitation when
65	judging stale placeholders or field caches.
66	- For `--qa deep` or `--qa strict`, prefer repairing/installing renderers before
67	generation. If the environment cannot run them, `deep` generates a degraded
68	manifest and `strict` fails with a visual proof blocker.
69
70	## Agent Workflow
71
72	1. Run the dependency preflight above and report any degraded capability.
73	2. Determine the brand name and locate the user-provided `.docx` template.
74	3. If no matc

People who liked this also viewed

6 picks

Type