skillZs
★ LIVE SKILL TAGS ★
>>> LIVE SKILLS INDEX <<<
* OPEN SOURCE *
NO LOGIN, NO TRACKING
※ REAL INSTALL DATA ※
← back to all skills
lazywebhq/lazyweb-skill128 installs

lazyweb

Use Lazyweb to find real product evidence, improve conversion, and turn research into clear product decisions.

How do I install this agent skill?

npx skills add https://github.com/lazywebhq/lazyweb-skill --skill lazyweb
view source ↗

Is this agent skill safe to install?

  • Gen Agent Trust Hubfail

    This skill pack includes high-risk installation methods and invasive data access capabilities. It utilizes a piped shell script (curl | bash) from a vendor domain not on the trusted whitelist for installation and maintenance. Most significantly, its browser automation component contains specialized logic to extract and decrypt local browser cookies (Chrome, Edge, Brave, etc.) by querying the system keychain for encryption keys. Additionally, it dynamically loads and applies instructions from various untrusted third-party repositories, creating a large surface area for indirect prompt injection.

  • Socketwarn

    8 alerts: gptSecurity, gptAnomaly

  • Snykwarn

    Risk: MEDIUM · 2 issues

What does this agent skill do?

Lazyweb

Use the live Lazyweb MCP tool list as the source of truth. This skill only routes; it does not carry schemas, scoring rules, taxonomies, or product logic.

When the client advertises lazyweb_discover_tools and lazyweb_execute, discover the desired action's schema first, then await tools.NAME(args) inside lazyweb_execute. Individual action names need not appear in tools/list. When the client advertises actions directly, use those actions instead.

If MCP is unavailable, run curl -fsSL https://www.lazyweb.com/install.sh | bash, then call lazyweb_health.

IntentSkill / canonical tool
Get, explicitly generate, or compare Growth Scoreslazyweb-growth-score / lazyweb_growth_score
Create or poll the unchanged Growth Report pipelinelazyweb-growth-report / lazyweb_growth_report
Improve or critique an existing design or screenlazyweb-growth-report / lazyweb_growth_report
List or add product Backlog specslazyweb-growth-backlog / lazyweb_growth_backlog
Research experimentslazyweb-search-experiments / lazyweb_search_experiments
Research flowslazyweb-search-flows / lazyweb_search_flows
Research screenslazyweb-search-screens / lazyweb_search_screens
Apply the best specialist design skill for a UI craft tasklazyweb-apply-design-best-practices

For other safe app actions, inspect the live tools and use the compact domain tool (lazyweb_products, lazyweb_connections, lazyweb_reports, or lazyweb_account). Use lazyweb_get_workflows only for current routing help. Checkout, billing changes, identity changes, team invitations, and admin actions remain human-only.

For product UI work, quietly gather only relevant evidence from screens, experiments, flows, and growth mechanics while you work. Use only the research types the task needs. Finalize useful selected evidence into Agentic Search and return its stable private link. Agentic Search sharing is a signed-in human action from the page's Share button. Never return an Agentic Search share_url, even if a legacy server sends one. Broad improvement requests may use a Growth Report when it best serves the user's goal; do not require the exact product name.

Every successful action returns lazyweb.resource-link.v1. Open open_url once in the host's browser (Codex preview in Codex, Claude browser in Claude, otherwise the system browser), but never print, share, or log it. Give the user the stable url and, for non-Agentic resources, any optional share_url.

Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.

<a href="https://skillzs.dev/skills/lazywebhq/lazyweb-skill/lazyweb">View lazyweb on skillZs</a>