# cv.ktortolini.dev

Public CV and resume source directory for Kevin Tortolini.

## Authentication

Private application and JSON API requests require a fresh signed X-Agent-ID-Token from https://auth.agentid.com for the registered Workdoc client ID. Refresh before expiry, send only to the canonical CV origin, and refuse redirects. Registered public discovery documents and /api/health remain readable without identity. An anonymous HTML root shows the Agent ID access notice.

## Primary Reading Paths

- [CV directory](https://cv.ktortolini.dev/?directory=local): Human-facing resume directory with links to the Markdown resume sections. Fetch https://cv.ktortolini.dev/index.md when Markdown is preferred.
- [Resume API](https://cv.ktortolini.dev/api/resume): JSON API containing the public Markdown resume source files and metadata.
- [Health API](https://cv.ktortolini.dev/api/health): JSON health check for the Workdoc CV application.

## Full Context

- [Bounded resume export](https://cv.ktortolini.dev/llms-full.txt): Fetch when a single Markdown document with all public resume sections is useful.

## Browser Automation

- Prefer semantic queries first: use the page heading named `/`, `/cv`, or `/404`, then activate directory rows by their button names.
- Stable fallback selectors use `data-testid` with `workdoc-static-foldkit-root`, `workdoc-directory-*`, `workdoc-shell-*`, `workdoc-isometric-background`, and `workdoc-glyph-sdf-canvas` anchors.
- Repeated rows expose `data-workdoc-row-kind` and `data-workdoc-row-index`; visual layers expose `data-workdoc-agent-region` and `data-workdoc-agent-role` without embedding private payloads.

## Capability Discovery

- [Agent Skills index](https://cv.ktortolini.dev/.well-known/agent-skills/index.json): Fetch when an agent wants a portable skill for reading the public CV endpoints.