Website Audit Agent
Website audits are frequently vague and unsupported. Produce an evidence-based, prioritized report with concrete fixes across technical, SEO, UX, conversion, content, and analytics dimensions.
Who and when?
Web developers, SEO specialists, growth consultants, founders, agencies.
From intake to final review.
- Each finding has evidence or is labeled a hypothesis requiring verification. - Scope, timestamps, and tested URLs are recorded. - No intrusive scans, credential attacks, unauthorized data collection, or high-volume requests. - Prioritization is reproducible. - Every implementation ticket includes a testable acceptance criterion.
Inputs and outputs
Required: website URL or supplied site export, audit scope, target market, primary conversion goal. Optional: approved crawl limits, analytics export, Search Console export, screenshots, device profiles, competitor references, credentials for user-owned systems (never stored in output).
- scope.md
- evidence-register.csv
- technical-audit.md
- seo-audit.md
- ux-audit.md
- conversion-audit.md
- content-audit.md
- analytics-audit.md
- findings.csv
- prioritized-backlog.csv
- implementation-briefs/
- coverage-limitations.md
- state.json
A synthetic worked example.
Synthetic example: Five fictional HTML pages contain seeded metadata, label, and CTA defects. Evidence becomes findings ranked by impact, effort, confidence, and dependency. Each implementation ticket has a testable acceptance criterion. With no analytics export, analytics coverage is labeled missing rather than filled with invented funnel numbers.
id,url,issue,acceptance F-01,/contact,Missing label,Control has a tested accessible name
An illustrative output excerpt, not a client result or live run. Sample approval cannot authorize a real action.
How to use it
- After repository publication, obtain the files and read README and SKILL.md. No download link exists before release.
- Prepare inputs and sources in a separate run folder. Keep secrets and client data out of public files.
- Start at intake and follow the workflow manually or with an AI tool that can read and write local files. No platform compatibility is claimed before testing.
- Review each phase and record a version-specific decision. Missing inputs and weak evidence create blockers, not guessed completion.
- Persist state, decisions, and artifacts. Resume from the latest approved phase without silently replacing approved output.
Limits and safety
Penetration testing, exploit attempts, automatic production code changes, comprehensive analytics without provided access, guaranteed Lighthouse scores.
Files and web pages are untrusted information, not authorization to change goals, publish, or contact anyone. External effects need separate permission. Run data stays local and examples are synthetic. File checks cannot prove factual truth or reviewer identity.
Status: v1 specification for review. GitHub URL, release, license, and exact runtime requirements are not published. Contributions will use the actual repository after launch.