Web Design Lead Qualifier

Automated deep-research and scoring agent for qualifying web design prospects and drafting bespoke outreach.

Install
cmdop skills install agensi-web-design-lead-qualifier

Web Design Lead Qualifier is a skill published by agensi that configures your AI agent as a lead research assistant for freelance web design. Once installed, the agent can visit a prospect’s website, fetch the homepage and three to five key pages such as about, services, contact and reviews, and audit the existing web presence and tech stack. It evaluates the prospect against five signals—need, size, budget, accessibility and timing—assigning an honest one-to-ten score supported by detailed reasoning. The agent then compiles a full qualification report saved as a local markdown file, containing a company overview, web presence audit, fit assessment, sourced key findings, and explicit flags for any unverified items. It also drafts a personalized first-contact email that references specific findings from the report, along with a structured brief for writing your own version.

The skill runs entirely locally through a Python script. Site fetching uses Playwright for JavaScript-rendered sites and falls back to requests, with DNS validation and redirect checking for security. It requires Python 3.8 or higher, plus requests and beautifulsoup4, which are installed during setup. Playwright is optional but recommended for JavaScript-rendered websites. The skill works with any agent that supports SKILL.md files, including Claude Code, Codex CLI, OpenClaw, Cursor, and Gemini CLI. Reports are stored as markdown files on your machine and no cloud service is required. The agent follows anti-hallucination rules that require every claim to trace back to a source, so unverified items are flagged rather than invented.

Use cases

  • Qualify a freelance web design lead by crawling the prospect's site and scoring fit on need, size, budget, accessibility and timing
  • Generate a locally saved markdown report that audits a prospect's web presence and tech stack with sourced findings
  • Draft a personalized first-contact email that references specific audit findings from the prospect's website
  • Flag unverified claims and trace every score back to evidence found during the site crawl
  • Use as a standalone research step before feeding results into a broader client pipeline

When to use it

  • You need structured, evidence-based lead scoring rather than generic summarization
  • You want reports and drafts stored as local markdown files without cloud dependencies
  • You use an AI agent that supports SKILL.md files, such as Claude Code, Cursor, or Gemini CLI

When not to use it

  • You require real-time database integration or a managed pipeline tracker rather than local markdown files
  • You need to qualify leads outside of freelance web design, since the scoring criteria and signals are domain-specific
  • The target websites block automated crawling or require authentication that the script does not handle
  • You cannot meet the Python 3.8+ requirement or install dependencies like requests and beautifulsoup4