> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://respan.ai/docs/documentation/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://respan.ai/docs/_mcp/server.

# Get started

## Docs

- [Overview](https://respan.ai/docs/documentation/overview.md): Respan is the AI gateway with built-in observability, evals, and security testing.
- [Project settings](https://respan.ai/docs/documentation/admin/projects.md): Create and switch projects, and manage a project's name, ID, and data retention.
- [Members](https://respan.ai/docs/documentation/admin/collaborate.md): Invite teammates to a project and choose what each of them can do.
- [API keys](https://respan.ai/docs/documentation/admin/respan-api-keys.md): Create, scope, limit, and revoke the keys your apps use to call Respan.
- [Provider keys](https://respan.ai/docs/documentation/admin/llm-provider-keys.md): Use your own OpenAI, Anthropic, Azure, or other provider keys with the gateway.
- [Billing and credits](https://respan.ai/docs/documentation/admin/credits-billing.md): Add gateway credits, set up auto reload, and manage your plan and payments.
- [Data webhooks](https://respan.ai/docs/documentation/admin/webhooks.md): Send spans, traces, scores, and billing events to your own endpoint.
- [Run a red team campaign](https://respan.ai/docs/documentation/red-team/run-assessment.md): Attack your agent with adversarial messages and get a graded security report.
- [Review results and troubleshoot](https://respan.ai/docs/documentation/red-team/results.md): Read a campaign's report, work through its findings, and fix runs that stop early.
- [MCP](https://respan.ai/docs/documentation/mcp.md): Let Claude, ChatGPT, Cursor, Codex, and other AI tools work with your Respan data and docs.
- [CLI](https://respan.ai/docs/documentation/cli.md): Set up Respan in a project, trace your coding agent, and query your Respan data from the terminal.
- [Respan Agent](https://respan.ai/docs/documentation/agent.md): Ask questions about your LLM traffic, and let the agent set up prompts, monitors, and evaluations for you.