sudoprivacy/sudowork/auto-login

Claude Code and agent skill by sudoprivacy/sudowork

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Log in to a website automatically using credentials saved in the local key store. Use when the user wants unattended/repeat login to a site (no re-scanning QR, no re-typing passwords), e.g. '帮我自动登录这个网站', 'set up auto-login for X', '记住这个网站的登录'. The plaintext password is NEVER seen by you — the user fills it in 秘钥管理 and sudowork fills the form itself.

Install

npx vskill@latest install sudoprivacy/sudowork/auto-login

SKILL.md

Website Auto-Login

Set up and run automatic website login. You explore the site and register the "login recipe" (which fields to fill); the user supplies the username/password in sudowork's 秘钥管理 UI; sudowork then fills the form (incl. image captcha) and submits — the password never enters your context.

All commands go through the helper (it talks to sudowork over the authenticated Auth Proxy you already have env for):

bash
python scripts/auto_login.py <list|register|start> [args]

Hard rule

NEVER ask the user to paste their password to you, and NEVER type a password into a page yourself. Credentials live only in the key store; you only handle the non-secret login recipe (selectors + url) and the entry title.

Workflow

1. Explore the login page (use the browser skill)

Open the login URL and inspect the form to find CSS selectors for:

  • the username/account input
  • the password input
  • the submit/login button
  • (if present) the captcha text input and its captcha `<img>` element

Verify each selector actually matches (e.g. browser_get_dom_snapshot / document.querySelector(...)). Note whether it's single_step (all on one page) or two_step.

2. Register the login recipe (non-secret)
bash
python scripts/auto_login.py register \
--title "<short name, e.g. yidao>" \
--url "<login page url>" \
--username-selector "<css>" \
--password-selector "<css>" \
--submit-selector "<css>" \
[--captcha-selector "<css>" --captcha-image-selector "<css>"] \
--strategy single_step
3. Ask the user to fill credentials (you stop here)

Tell the user to open 设置 → 远程连接 → 秘钥管理 → 网站自动登录, expand the entry with the title you used, fill 用户名 + 密码, and click 保存凭证. Then ask them to tell you when done. Do not proceed until they confirm.

4. Run the auto-login
bash
python scripts/auto_login.py start --title "<title>"

sudowork reads the credential from the key store, fills the form in the running browser (solving any image captcha with a vision model), and submits. The result reports success/navigation only — never the password.

Check status anytime
bash
python scripts/auto_login.py list # sites + whether each already has a saved credential

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
10
Repository forks
2
7-day trend score
17

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/sudoprivacy/sudowork/auto-login/badge)](https://verified-skill.com/skills/sudoprivacy/sudowork/auto-login)

Something wrong with this skill or its page? Tell us.