Install
Get Started
Install the log10x MCP server into your AI client, then ask it to run a cost POC. No account or API key needed to start: it boots in demo mode against a live EKS demo.
Claude Code:
Claude Desktop, Settings → Developer → Edit Config:
Add to Cursor, or paste into ~/.cursor/mcp.json:
Add to ~/.codeium/windsurf/mcp_config.json:
Add to .vscode/mcp.json:
Add to .vscode/mcp.json (JetBrains: ~/.config/github-copilot/intellij/mcp.json):
Run the server and wire it into any MCP-compatible client:
Then ask your AI client cost POC on the log10x demo to try the live demo, or cost POC on our Splunk against your own stack.
To connect your own data, ask the agent to sign in to Log10x (GitHub device flow, no key to paste) or set LOG10X_API_KEY. Per-client config locations, credentials, and tool-discovery tips: MCP Server install.
Engine
The 10x engine is the runtime every app runs on. Install it directly with a container image, a single-line script, or an OS package below; the apps also bundle it through their Helm charts.
No licence is needed to evaluate: with none configured, the engine runs the full product on its built-in 30-day evaluation licence (10 nodes, air-gapped). A production licence (see licensing) is passed as TENX_LICENSE_FILE (a mounted file) or TENX_LICENSE_KEY (the JWT inline).
Container Deployment
| Target | Installer | Flavors () |
|---|---|---|
| Kubernetes | Helm Charts | |
| Docker | Docker Image |
Single Line Installer
| Target | Installer | Flavors () |
|---|---|---|
| Linux (most distros) | Single line script | |
| macOS | Homebrew (brew install log-10x/tap/log10x) |
|
| Windows | PowerShell script (irm ... \| iex) |
Windows has no native Runtime: no tenx-*-windows-*-native asset is built. The Windows installers ship Runtime (JVM), the same run-only build with a bundled JRE, and the Compiler.
Download
| Target | Installer | Flavors () |
|---|---|---|
| Ubuntu | DEB package / Native Binary | |
| Debian | DEB package | |
| Red Hat / CentOS 7 | RPM package | |
| Red Hat / CentOS 8+ | RPM package | |
| Windows | MSI package (manual) | |
| macOS | Homebrew (brew install log-10x/tap/log10x) |
Metrics in Datadog
A Datadog Agent check reads the engine's Prometheus endpoint and submits the per-pattern counters to Datadog: what the pipeline read, what it forwarded, the encoded size when compaction is on, and what went to offload storage and came back by query. The Agent scrapes the engine, so no Datadog credential is configured on the engine.
After installing, ask your AI client for a cost POC and the engine returns your first analysis in 5 minutes.