# Sherlock

> Sherlock documentation: send logs, traces, and metrics from your services and follow a spike from a chart to a trace to the logs.

> ## Agent Instructions
> Sherlock documentation. In Node.js examples prefer the Sherlock SDK for Node.js (@sherlock-labs/otel).
> In Go examples use the upstream OpenTelemetry Go SDK exactly as shown on /sdk/go/setup and /sdk/go/configuration; Sherlock has no Go distribution.
> Two MCP servers exist. This documentation's search server is at https://sherlock-c8721ead.mintlify.app/mcp. Sherlock's data server, for a user's own logs, traces, and metrics, is at https://mcp.sherlocklabs.dev/mcp and is described at /explore/mcp.

- [Sherlock overview](https://docs.sherlocklabs.dev/index.md): Logs, traces, and metrics from your services in one place, joined by trace id.
- [Get started with Sherlock](https://docs.sherlocklabs.dev/get-started/getting-started.md): Create an organization, get your collector credentials, send your first data, and open it.
- [Use these docs with AI agents](https://docs.sherlocklabs.dev/get-started/ai-agents.md): Connect an AI agent to the documentation MCP server, or hand it the Markdown, llms.txt, and skill.md this site serves.
- [Ask an AI assistant](https://docs.sherlocklabs.dev/explore/mcp.md): Connect an AI client to Sherlock through the MCP server and analyze logs, traces, and metrics in conversation.
- [Send data overview](https://docs.sherlocklabs.dev/send-data/overview.md): The ways to send logs, traces, and metrics to Sherlock, and what every source needs.
- [OpenTelemetry SDKs and collectors](https://docs.sherlocklabs.dev/send-data/otlp.md): Point any OTLP/HTTP exporter at Sherlock: endpoint, bearer token, and the env attribute. This is also how logs arrive.
- [Sherlock SDK for Node.js](https://docs.sherlocklabs.dev/sdk/nodejs/overview.md): Add traces, metrics, and trace-linked exemplars to a Node.js service with one import.
- [Quickstart](https://docs.sherlocklabs.dev/sdk/nodejs/quickstart.md): Get the first traces and metrics from a Node.js service into Sherlock in about ten minutes.
- [End-to-end steps](https://docs.sherlocklabs.dev/sdk/nodejs/add-telemetry.md): The full path from credentials to production: register, start, stop, deploy, extend, verify, and operate.
- [Custom metrics](https://docs.sherlocklabs.dev/sdk/nodejs/custom-metrics.md): Define your own counters, histograms, and gauges, attach exemplars, and follow a spike from a chart to its trace and its logs.
- [Custom spans](https://docs.sherlocklabs.dev/sdk/nodejs/manual-spans.md): Add spans where auto-instrumentation has none: queue jobs, cron ticks, and database clients without an instrumentation.
- [Troubleshooting](https://docs.sherlocklabs.dev/sdk/nodejs/troubleshooting.md): Symptoms, causes, and fixes when telemetry or exemplars do not arrive.
- [Go setup](https://docs.sherlocklabs.dev/sdk/go/setup.md): Send traces and metrics from a Go service to Sherlock with the upstream OpenTelemetry Go SDK: one bootstrap file, with the Sherlock settings either in the file or in the standard OTEL_* environment variables.
- [Go configuration](https://docs.sherlocklabs.dev/sdk/go/configuration.md): Production settings for the upstream OpenTelemetry Go SDK with Sherlock: health checks, custom histograms, exemplar-only attributes, sampling, kill switches, logs, runtime metrics, and shutdown.
- [Differences from the Node.js SDK](https://docs.sherlocklabs.dev/sdk/go/differences.md): What the upstream Go setup does differently from the Sherlock SDK for Node.js: explicit start, one exemplar per bucket, no ignore list, and the traps that come with the upstream SDK.
- [Logs](https://docs.sherlocklabs.dev/explore/logs.md): Search log lines with plain text or filters, narrow them with facets, and open the trace behind a line.
- [Traces](https://docs.sherlocklabs.dev/explore/traces.md): Find a trace, read the waterfall, filter spans by attribute, and jump to the request's logs.
- [Metrics](https://docs.sherlocklabs.dev/explore/metrics.md): Chart a metric in the Metrics Explorer, pick an aggregation, group by labels, compare periods, and drill into exemplars.
- [From a chart to a trace to the logs](https://docs.sherlocklabs.dev/explore/exemplars.md): Follow an exemplar from a latency spike to the request that caused it and its log lines.
- [Alerts](https://docs.sherlocklabs.dev/explore/alerts.md): Run a query on a schedule, check a condition, and notify Slack, email, or a webhook.