Index of every page on this site: llms.txt
All skills
e2e-inc avatar

/create-verification-skill

@a5d1788 #9 trending · Oct 2026
by e2ee2e-inc/e2e14k stars
782
Run it yourself

Generate a project-local verify-<app> skill on top of e2e, so any coding agent can launch the app, check it is worth driving, exercise a feature the way a user does, keep evidence, and bug bash it. Writes the skill, a feature map with one explore charter per feature, and proves it once. Use for /create-verification-skill, "make a verification skill for this repo", or when a project has no scripted way to prove UI behavior.

Use this Skill: https://skilld.dev/gh/e2e-inc/e2e/create-verification-skill

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

referencesexampleverify-playgroundfeaturestodos.md

≈458 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Todos

A todo list kept in the browser: add, complete, filter, delete, with a remaining count.

Sub-features

  • Add with the Add button or Enter in the New todo field; an empty submission adds nothing.
  • Complete with the checkbox labelled by the todo's title.
  • Filter with the Filter tablist: All, Open, Done.
  • Delete with the Delete <title> button; the list persists across reloads.

How to get to it (user POV)

Nav link Todos, or /todos. No account.

Driving it with e2e

tests/todos.e2e.ts pins all of it. Locators, checked with locate: screen.getByLabel('New todo'), screen.getByRole('button', 'Add'), screen.getByTestId('todo'), screen.getByRole('status', 'Remaining'), screen.getByRole('tablist', 'Filter').getByRole('tab', 'Open'), screen.getByLabel('<title>') for a todo's checkbox, screen.getByRole('button', 'Delete <title>'). Live walk: open_session, navigate /todos, type into the New todo node, press Enter, observe.

Proof

On screen: one todo item per title in order, Remaining reads <n> remaining, the selected tab hides the others' items. Store: localStorage.todos holds the same titles and done flags (browser.evaluate(() => localStorage.getItem('todos'))).

Charters

todos-flow|web|skeptic|Starting at /todos, add, complete, filter, and delete todos; check the remaining count and every filter tab against the list
todos-input|web|fuzzer|Starting at /todos, submit the new todo field empty, with 300 characters, with unicode, and with leading spaces; report what is accepted, trimmed, or lost

Gotchas

  • Within one session or explore run, todos survive navigation and reload by design. Every test and every open_session starts from a fresh browser context, so a list that is not empty at the start of one is a bug.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

1 warning4d3 checks · Risk SAFE
  • Gen Agent Trust Hub4d

    This skill is used to generate sub-skills for application verification. It includes functionality to execute project-specific commands and generate new scripts, which is expected for developer tools. The primary security consideration is a low risk of indirect prompt injection, as the skill processes untrusted local code and documentation to generate instructions for subsequent agent actions.

  • Socket4d

    No alerts

  • Snyk4d

    Risk: MEDIUM · 2 issues

Signed by skilld at a5d1788. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 5 days ago
disable-model-invocation
true
Other metadata
metadata
{
  "openclaw": {
    "homepage": "https://e2e.com",
    "requires": {
      "bins": [
        "node",
        "npx"
      ]
    }
  }
}

README badge

README badge for e2e-inc/e2e/create-verification-skill