replay-mcp

Inspect Replay.io recordings to answer questions about web application behavior.

41|4|Updated May 6, 2026
One-click install
npx skills add https://github.com/markerikson/opencode-config-example --skill replay-mcp-markerikson
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: replay-mcp
Source: https://github.com/markerikson/opencode-config-example/tree/main/config/skill/replay-mcp
Command: npx skills add https://github.com/markerikson/opencode-config-example --skill replay-mcp-markerikson

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Replay MCP helps you answer precise questions about what a web app did during a specific user session by inspecting the recorded execution timeline rather than guessing from symptoms.

Core Features & Use Cases

  • Record-driven investigation: Explore errors, timeline events, console messages, local storage access, network requests, and React renders to understand what happened.
  • Causality and explanation: Use stack and control-dependency views to determine why particular points executed (or failed to execute).
  • State and source forensics: Inspect variables, DOM element details, React component trees, logpoints, and evaluate expressions at specific points to connect behavior to underlying code.

Quick Start

Ask the AI to use the replay-mcp tools to find the first network request and related error responsible for a failing button click in a Replay recording.

Frequently Asked Questions about replay-mcp

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I debug a broken React UI flow using a recorded web app session?

Debug a broken React UI flow by inspecting the recorded execution timeline to analyze React renders, console messages, and network requests with evidence from specific execution points.

What is the best way to find the network request responsible for a client-side failure?

Find the failing network request by exploring timeline events and stack dependencies in the recorded session to trace the exact execution point where the client-side failure originated.

Why does a specific code point execute or fail to execute during a web app investigation?

Determine why a code point executed or failed by using stack and control-dependency views from the recorded timeline to establish causality and connect behavior to the underlying source.

Can I inspect variables and evaluate expressions at a specific point in a recorded session?

Inspect variables, DOM element details, and React component trees, or evaluate expressions at specific execution points to perform state and source forensics on the recorded timeline.

Do I need a replay.io recording to analyze timeline forensics for my web application?

You need a web application session recorded on replay.io to apply timeline forensics, as the investigation requires exploring execution points, sources, and dependencies from that recorded data.

What are the limitations of debugging web app behavior from recorded execution timelines?

Timeline forensics are limited to analyzing interactions captured during the specific recorded session, meaning unrecorded user flows or external state changes outside the recording cannot be investigated.