> ## Documentation Index
> Fetch the complete documentation index at: https://docs.foggyhq.com/llms.txt
> Use this file to discover all available pages before exploring further.

# FAQ

> Frequently asked questions about Foggy — setup, security, models, pricing, and more.

## Getting Started

<AccordionGroup>
  <Accordion title="How long does setup take?">
    Under 5 minutes. Sign up, add a connector (URL + API token), and ask your first question. See the [Quick Start](/overview/quickstart) guide.
  </Accordion>

  <Accordion title="Does Foggy replace Grafana / Datadog / my existing tools?">
    No. Foggy reads from your existing tools — it doesn't replace them. It works **on top of** your current observability stack and queries Grafana, Prometheus, and others to answer your questions.
  </Accordion>

  <Accordion title="What integrations are supported?">
    Observability: Grafana, Prometheus, OpenSearch, Kubernetes. Errors and issues: Sentry, GitHub, Linear. Full-stack observability: Datadog. Alerts: generic Alert Webhook (any alerting tool that can POST JSON). Chat: Slack. Private-network proxy: Satellite. See the **Integrations** section in the sidebar for setup guides for each.
  </Accordion>
</AccordionGroup>

## AI & Models

<AccordionGroup>
  <Accordion title="What LLM models does Foggy use?">
    Claude Sonnet 4.6 (default), Claude Opus 4.7, and other models. You choose per-message via the model selector in the chat input.
  </Accordion>

  <Accordion title="Is my data sent to LLM providers?">
    Investigation prompts and tool query results (metrics, logs, events) are sent to the selected LLM for analysis. **Infrastructure credentials (API tokens, URLs) are never sent to LLMs.** See [Security](/platform/security#llm-data-flow) for the full data flow.
  </Accordion>

  <Accordion title="How long does an investigation take?">
    Simple queries: 10–30 seconds. Complex multi-tool investigations: 1–3 minutes. Results stream in real-time.
  </Accordion>

  <Accordion title="Can I use my own LLM API keys?">
    Yes. Paste your own LLM API key in Settings → LLM Providers. You control which provider processes your data.
  </Accordion>
</AccordionGroup>

## Security & Data

<AccordionGroup>
  <Accordion title="How does Foggy handle security and compliance?">
    Foggy operates in read-only mode with least-privilege access. See the [Security](/platform/security) page for full details.
  </Accordion>

  <Accordion title="Does Foggy modify my infrastructure?">
    Never. Foggy uses **read-only access** to all connected tools. It queries dashboards, metrics, and logs but never creates, updates, or deletes resources in your infrastructure.
  </Accordion>

  <Accordion title="How are credentials stored?">
    Encrypted at rest using Fernet symmetric encryption. Decrypted only when making API calls to your tools. Never logged, cached, or displayed in the UI after initial entry.
  </Accordion>

  <Accordion title="Do you store or reuse our production data?">
    No. Foggy does not store raw production data or use it to train models. Data is retained only as conversation history within your project and is never shared across projects.
  </Accordion>
</AccordionGroup>