Lightrun Ai MCP Server

A collection of reusable skills for investigating runtime issues with Lightrun. That is what the lightrun ai mcp server brings to an AI assistant: the same capability, reachable through the Model Context Protocol rather than a separate app or dashboard.

The tools it exposes

The server publishes 7 tools. What each one is for:

• lightrun-live-runtime-debugging — deterministic live-runtime investigation workflow with preflight checks, evidence capture, and handoff output
• lightrun-error-remediation-automation — error remediation automation workflow with root cause analysis, mitigation strategies, and automated fixes
• lightrun-ask-prod — live production runtime question answering for current values, durations, hit counts, and distributions
• lightrun-runtime-aware-pr-review — pull request review workflow that combines live runtime samples with diff-aware risk analysis
• Cursor — 1. Open Cursor. 2. Add plugin from repository URL (or local path). 3. Use this repository URL: https://github.com/lightrun-platform/lightrun-ai
• Claude — Use your Claude plugin/marketplace flow to install from this repository
• Codex — 1. Copy or symlink skill folders from skills/ into one of: - $REPO_ROOT/.agents/skills - $HOME/.agents/skills 2. Restart/reload Codex if needed

What it needs from you

• Access to Lightrun MCP server - Completed MCP OAuth authorization for your environment

Getting it running

Setup follows the usual MCP pattern — install or clone the server, register it in your client's configuration file, restart the client.

How it compares

This sits in the developer tooling group, where several servers overlap in what they claim to do but differ sharply once you actually set them up. Lightrun Ai's toolset — lightrun-live-runtime-debugging, lightrun-error-remediation-automation, lightrun-ask-prod and 4 more — is a fair guide to whether it matches your workflow. It is maintained by lightrun-platform; worth a glance at recent repository activity before you build anything load-bearing on it.

We check each listing at SyncDev against the project's documentation before it goes live — if something here drifts out of date, it is a bug worth reporting.

Things to watch

• It runs with your machine's permissions. That is convenient and also the reason to think about what you point it at before you approve a tool call.
• Keep per-call confirmation enabled while you learn its behaviour; it is the cheapest safeguard you have.