Vetrix Docs

Vetrix LLM documentation

This is the LLM-optimized documentation set for the Vetrix platform at gitvetrix.com. It is written for AI assistants and other automated readers working against the API, the CI/CD system, and a running server, so its pages lead with exact endpoint shapes, configuration keys, file paths, and command sequences rather than with screen-by-screen instructions.

It is a separate set from the human-facing user documentation, which lives in the user-docs directory at the root of this repository and walks a person through the Vetrix web UI in the browser. The two sets are maintained independently and do not link to each other, so choose the one that matches your reader. The user-docs set listed below sits inside this tree and is the LLM-optimized end-user set; it is not the repo-root directory of the same name.

Documentation sets

  • Admin documentation — operator and administrator reference for a self-hosted instance, covering the admin surface and runtime configuration, git storage and mirrors, CI/CD runners and pipeline variables, the package registries, the email template system, and backup, search-indexing, and upgrade maintenance.
  • API documentation — reference for the HTTP API, the OCI, npm, PyPI, and Go module registry endpoints, and the machine-readable OpenAPI specifications, including the common error envelope and how request rate limits are scoped and enforced.
  • Runbook documentation — task-oriented operator procedures for CI/CD, search indexing, storage and backups, database migrations, and optional platform features: how to operate each subsystem, what to check when it misbehaves, and how to recover it.
  • System documentation — system, architecture, and internal-design reference, covering Docker, Kubernetes, and nginx deployment, CI/CD, search, frontend, and security internals, the contributor testing rules, accessibility verification, and the code-scanning tooling configuration.
  • User documentation — end-user guides for authentication, repositories and the issue tracker, CI/CD pipelines, package registries, and the webhook, MCP, and code-search integrations, starting from a getting-started page.

Each set carries its own README.md index and its own doc-lint.sh, which enforces the operator-manual rules for that set.