# Tracekit product overview

Tracekit helps software teams find what broke in production. It keeps telemetry and runtime evidence connected to the same request path.

## What Tracekit connects

- [Product Analytics](https://tracekit.dev/features/analytics.md): See live visitors, traffic sources, goals, Stripe revenue, and the related traces or session replays.
- [Distributed Tracing](https://tracekit.dev/features/distributed-tracing.md): OpenTelemetry-native tracing across every service. See the full request lifecycle from browser to database.
- [Error Tracking](https://tracekit.dev/features/error-tracking.md): Errors with full trace context. See what happened before, during, and after every exception.
- [GitHub Pull Request Remediation](https://tracekit.dev/features/github-remediation.md): Turn anomaly or error-trace evidence into a reviewable pull request when Tracekit finds a useful code fix.
- [Session Replay](https://tracekit.dev/features/session-replay.md): Replay user sessions linked to traces and errors. See exactly what the user experienced.
- [Dynamic Logs](https://tracekit.dev/features/live-debugging.md): Add dynamic logs in production without redeploying. Capture variable state, stack traces, and context on live traffic.
- [LLM Observability](https://tracekit.dev/features/llm-observability.md): Monitor LLM calls across providers. Track latency, token usage, costs, and prompt/response quality.
- [Alerting & Incidents](https://tracekit.dev/features/alerting.md): Smart alerts with Slack, Discord, Teams, PagerDuty, and OpsGenie. Triage inbox with AI-powered insights.
- [AI MCP Server](https://tracekit.dev/features/ai-mcp-server.md): Give Claude and compatible AI assistants secure access to live production context and supporting evidence.

## How teams use Tracekit

1. Send OpenTelemetry traces, errors, browser sessions, analytics events, and LLM spans.
2. Follow a request across services, releases, database calls, and user sessions.
3. Add dynamic logs to capture runtime state without another deployment.
4. Ask Tracekit Copilot or an MCP-compatible assistant for an evidence-backed answer.
5. Open the related trace, replay, alert, capture point, or code owner.

## Start

- [Read the documentation](https://tracekit.dev/docs/index.md)
- [Review pricing](https://tracekit.dev/pricing.md)
- [Create a free account](https://app.tracekit.dev/register)