hos-user-manual

Generate end-user HTML operation manuals by driving a live E2E environment and capturing screenshots.

32|Updated Aug 29, 2026
One-click install
npx skills add https://github.com/openreachtech/hora-skills-ort-support --skill hos-user-manual-openreachtech
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: hos-user-manual
Source: https://github.com/openreachtech/hora-skills-ort-support/tree/main/kit/skills/hos-user-manual
Command: npx skills add https://github.com/openreachtech/hora-skills-ort-support --skill hos-user-manual-openreachtech

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Writing end-user manuals from source code produces inaccurate, screenshot-free documents that drift from the real product. This Skill generates operation manuals grounded in actual UI behavior by running the system for real, walking each feature, and capturing every screen. ## Core Features & Use Cases - Operation-based authoring: Starts the local E2E environment, logs in as a seeded test user, and performs every documented step through the UI so the manual reflects real behavior. - Structured HTML output: Produces one HTML page per feature with numbered steps and screenshots, plus a required table-of-contents page, all styled by a shared self-contained stylesheet. - Version-bound output: Writes each manual set under docs/manuals/v<version>/ tied to the repository's package.json version, preserving old versions as immutable history. - Use Case: After releasing version 0.3.0 of a web app, invoke this Skill to walk the login, registration, and admin features in the E2E stack and publish a complete Japanese-language user manual with screenshots under docs/manuals/v0.3.0/. ## Quick Start Generate the end-user operation manual for the current product version by walking every feature in the running E2E environment and writing the HTML pages with screenshots.

Frequently Asked Questions about hos-user-manual

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate a user manual with screenshots automatically?

This Skill starts the local E2E environment, logs in as a seeded test user, walks each feature through the UI, and captures one screenshot per meaningful step. It then writes one HTML page per feature plus a table-of-contents page under docs/manuals/v<version>/.

What prerequisites are needed to create an operation manual from a running system?

A working E2E environment with up, seed, and teardown commands must already exist in the repository. If it is missing, the Skill stops and directs you to build that environment first, since all steps and screenshots must come from real operation.

Can I write manual steps from source code instead of running the app?

No. The Skill enforces an operation-based rule: every step must be actually performed in the UI and every screenshot actually captured during the run. Steps that cannot be performed are reported as gaps, never invented.

How are manuals versioned when the product releases a new version?

Each manual set is bound to the repository's package.json version and written to docs/manuals/v<version>/. A new version gets a fresh directory with newly captured screenshots; past version directories are never edited.

What happens when a feature cannot be walked during manual generation?

The feature's page is omitted or the failing step is marked unavailable, and the issue is reported to the user as a finding. The Skill never fabricates steps or silently drops features without disclosure.