LightKeeper MCP
LightKeeper MCP lets developers run regional Lighthouse audits and generate performance fixes directly within AI coding agents.
Open productWhat it does
About LightKeeper MCP
LightKeeper MCP is a web performance audit integration created by sole developer Denys Filchakov. Designed to connect with AI coding tools such as Codex and Claude Code, it enables developers to initiate multi-region Lighthouse audits and inspect detailed browser evidence directly inside their coding workflow. The system collects metrics including Performance score, FCP, LCP, TBT, CLS, and Speed Index alongside diagnostic artifacts like JSON/HTML reports, HAR files, Performance traces, and DevTools protocol logs. Using built-in prompts like analyze_lighthouse_audit and plan_lighthouse_improvements, users can turn audit results into code fixes. Backend operations run on AWS and GCP using serverless architecture and Postgres task polling.
- Availability:
- unknown
- Access:
- unknown
- Format:
- software
Core capabilities
Regional Lighthouse Auditing
Executes Lighthouse performance audits across different geographic regions from inside AI coding workspace sessions.
Browser Evidence Artifact Collection
Collects detailed browser diagnostic files including Lighthouse JSON and HTML reports, HAR data, Performance traces, and DevTools protocol logs.
Keep discovering
Related debuts
Continue with products that solve a similar problem or serve the same kind of maker.
