# Garnet — Runtime visibility & control for untrusted code

> Detect and block unexpected outbound network calls from dependencies, MCPs, and AI-generated code during runtime in automation pipelines and Kubernetes.

## Product
- Garnet profiles runtime behavior (process lineage, network egress, file access) using a kernel-level eBPF agent (Jibril), exposed as Run Profiles and assertions in GitHub workflows.
- Primary surfaces: pull-request feedback, Slack, and the Garnet web app.

## Official URLs
- Marketing site: https://garnet.ai/
- Documentation: https://docs.garnet.ai
- App: https://app.garnet.ai
- Jibril (agent): https://jibril.garnet.ai
- Resources & articles (canonical; long-form MDX): https://garnet.ai/resources
- Legacy /blog URLs permanently redirect to /resources (same slug under /resources/[slug]).
- Pricing: https://garnet.ai/pricing
- Company: https://garnet.ai/about

## Install (GitHub Actions)
```yaml
- uses: garnet-org/action@v2
  with:
    api_token: ${{ secrets.GARNET_API_TOKEN }}
```

## Keywords
runtime visibility, eBPF, run profile, runtime assertions, supply chain security, AI-generated code, GitHub Actions, egress monitoring, behavioral detection

## Contact & social
- X: https://x.com/garnet_labs
- GitHub org: https://github.com/garnet-org
- LinkedIn: https://www.linkedin.com/company/garnet-labs

## Machine-readable discovery
- Sitemap: https://garnet.ai/sitemap.xml
- This file: https://garnet.ai/llms.txt