Borrowing it
Nothing to install: this file belongs to cardstack/boxel. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/cardstack/boxel/main/.claude/skills/secure-origin/SKILL.mdgit clone --depth 1 https://github.com/cardstack/boxelWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/cardstack/boxel/secure-origin)<a href="https://agentmods.dev/skills/cardstack/boxel/secure-origin"><img src="https://agentmods.dev/badge/skills/cardstack/boxel/secure-origin/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/cardstack/boxel/secure-origin"><img src="https://agentmods.dev/badge/skills/cardstack/boxel/secure-origin.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00289 | $0.05809 |
| Opus 5.5 | $0.00116 | $0.02324 |
| Sonnet 5.5 | $0.00058 | $0.01162 |
| Haiku 4.5 | $0.00029 | $0.00581 |
Grade A, and why
secure-origin scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
The secure origin
The governing fact: the secure origin is exactly as safe as the least trusted thing it serves. It exists so that a credential typed into it can't be read by card code. The browser's same-origin policy does the isolating: the app page, and every card evaluated in it, can't read another origin's document, keystrokes or form values. That guarantee is about the origin, not the page. One script served anywhere on the origin, from any path, can read every window open on it. So the origin serves the pages packages/secure-origin builds — the shell, and the credential fields the app frames — and nothing else. No realm, card module, realm asset, host app bundle or third-party script, ever.
What it serves
- The shell, at the path of the configured URL (
/deployed,/securein development): the builtindex.html, with no inline script, the files it loads (its entry pointshell.js; the chunks that imports, among themshared.jswith Ember and the components both entries use, the shell's own components staying inshell.js;layers.css, the order of the pages' cascade layers, which the build links ahead of every other stylesheet; and the components' stylesheets,shell.cssamong them withglobal.css's page-wide rules: boxel-ui's reset, the typeface and the page's background), and boxel-ui'svariables.css,theme.cssand the app's typeface, read from boxel-ui as they are.createSecureOriginbuilds aroutestable of every path served, from the build's manifest (see below). - Modes are added to
shell.js'smodestable, keyed by themodequery parameter. Every mode is the same shell document, with the same headers. They aresign-in,cli-sign-in(boxel keys sign-in) andcli-auth(boxel profile add, which hands the CLI a Matrix login token; see thesecure-sessionskill) (components insrc/shell/sign-in/, mounted withrenderComponent),registerandverify-email, where a registration email's link lands (src/shell/register/), andforgot-passwordandreset-password, where a reset email's link lands (src/shell/password-reset/); Google returns togoogle-callback, which serves the shell too.sign-outends the window's session and goes back to the app page in its#return=. An unknown mode renders nothing. A sign-in the app started carries#return=too, and the window hands it back with a login token (thesecure-sessionskill);returnis honoured only on an origin in the shell'sboxel-app-originsmarkup, which the realm server fills with the credential fields' app origins. - The pages are the app's sign-in screens, the same components in both places, and every one is a Glimmer component.
src/components/holds AuthContainer, AuthButton and AuthFormField and ports of the boxel-ui components they use (Button, FieldContainer, Label, BoxelInput, InputGroup, LoadingIndicator and the icons), each with boxel-ui's scoped CSS.src/screens/holds the screens themselves, presentational and stateless: sign-in, the registration steps and the password-reset steps, each taking its values and actions as arguments and its password fields as a block. The host renders its Login, RegisterUser and ForgotPassword from these, with its own controller (the Matrix service, the router) and the credential frame as the password fields (ADR 008 §1.9); the host may import from the package, and the package never imports from the host. On the secure origin,src/shell/sign-in/holds the sign-in form's controller, the CLI sign-in, the Google callback and the pages a sign-in ends on;src/shell/register/andsrc/shell/password-reset/hold account creation and password reset, with the host's RegisterUser and ForgotPassword rules, and render the same screens with the package's own inputs. The decisions behind them are plain modules: the CLI's loopback target and hand-off (src/security/cli-handoff.js), the registration's stages and the new account's first sign-in (src/security/registration.ts), and the reset link (src/shell/password-reset/reset-link.ts). The icon components copy boxel-ui's paths for the logo, Google's "G", the check mark and the spinner. A change to one of these screens' look or copy is made once, in the package, and reaches both origins. The CLI sign-in pages, the app's (cli-auth.gts) and the secure origin's (src/shell/sign-in/cli-auth-page.gts), render the sign-in screen too, naming themselves with the screen's@titleand saying what signing in grants the CLI in itsintroblock, whose type and spacing the screen sets. Each screen's controls carry both origins' test attributes (data-test-login-btnbesidedata-test-secure-password-submit, and so on), so keep both when editing one.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- today Changed a4acabbc16c3
- yesterday Changed · +1 lines 5cd9b7409a99
- 3d ago Changed · +15 lines · -3 tokens per session 1636fb0a08a4
- 5d ago First seen · 80 lines · 292 tokens per session scan A f6a36b1aa779
secure-origin is a skill published in the GitHub repository cardstack/boxel (27 stars, last pushed today), licensed MIT. It adds 289 tokens to every session and 5,809 once invoked, about $0.0012 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-21.
Other skills, from other repositories
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
component-fixtures
Use when creating or updating component fixtures for screenshot testing or their shared infrastructure, or when designing UI components to be fixture-friendly. Covers fixture file structure, theming, service setup, CSS scoping, async rendering, validation, and common pitfalls.