Vetrix user documentation

This is the human-facing guide to the Vetrix web UI at gitvetrix.com. It describes what you see and do in the browser, and it is deliberately separate from the LLM-optimized documentation set kept elsewhere in this repository: that set is written for AI assistants working against the API, while every page here is written for a person clicking through the site.

Each page opens with a short summary, tells you Where to find it, and then walks through tasks one at a time. UI labels appear in bold exactly as they are rendered on screen.

Account

Create an account, sign in, and manage your profile, credentials, and notification preferences. See the account guide for the full list of account pages.

Navigation

Find your way around: your newsfeed, discovery pages, search, your notifications inbox, and profiles. See the navigation guide for the full list of navigation pages.

Repositories

Create and browse repositories, collaborate on the work in them, and configure how they behave. See the repositories guide for the full list of repository pages.

CI/CD

Run pipelines, configure them, deploy to environments, and read the results. See the CI/CD guide for the full list of CI/CD pages.

Registry

Publish, browse, and consume packages stored alongside a repository. See the registry guide for the full list of registry pages.

Integrations

Connect a repository to the tools and services you already use. See the integrations guide for the full list of integration pages.

Administration

Run and configure a Vetrix instance. See the administration guide for the full list of admin pages.

Screenshots

screenshots/ holds the image assets referenced by the pages above.

Contributing — link and fragment checks

Before opening a pull request, run the two stdlib-only checkers in this directory, in addition to each set's doc-lint.sh:

python3 user-docs/check-links.py
python3 user-docs/check-fragments.py

check-links.py verifies that every relative link and image target in user-docs/, in llm-documentation/, and in every in-scope repository-root *.md file resolves to a file that exists, and that neither tree links into the other. The root scan is non-recursive (it lists the repository root directly; it never walks it, so .git/, user-docs/, and llm-documentation/ stay out) and covers every top-level *.md file except CHANGELOG.md, which is carved out by name in the script — see the ROOT_MD_EXCLUDE comment in check-links.py for why. Root-level files are the shared landing surface, so pointing into both trees is expected of them and the cross-tree rule is not applied to them. check-fragments.py scans both documentation trees and verifies that every #fragment on a link whose file target does resolve points at a real heading or <a name="..."> anchor in that file. Both exit non-zero on any violation.