Frequently asked questions
Whether you’re preparing for a repair, planning a build, or setting up your agent for the first time, this page answers the practical questions that tend to come up before you begin.
Setup
What do I need to use rx7db?
You need an MCP-compatible AI client and your own Anthropic API key. Your client must let you send a custom request header to the MCP server. Claude Code is supported; other MCP clients can connect when they support the same streamable-HTTP transport and headers.
Start with the command in the connection guide, then ask a tool in plain language about your car, symptoms, or build goal.
How do I connect in Claude Code?
Add the rx7db MCP endpoint and provide your key as the x-api-key header. The exact command and endpoint are kept up to date in the connection guide.
Why is my first answer taking a while?
Each request runs a focused research agent, so detailed answers can take a few minutes. If your client reports a timeout, increase its MCP tool timeout; Claude Code uses the MCP_TIMEOUT and MCP_TOOL_TIMEOUT environment variables.
Billing
How am I billed?
We don't charge for access to rx7db at the moment. However, you need to supply your own Anthropic API key, and usage is billed by Anthropic to the account that owns that key. We don't ask for your payment details or charge a separate subscription.
What does a question cost?
Expect roughly US$0.05–$0.50 per tool call. The actual cost depends on the question, how much research it needs, and the model available to your Anthropic account. Check your Anthropic usage dashboard for the authoritative total.
Can I control or limit my spend?
Yes. Manage spend limits, usage alerts, and key access in your Anthropic account. For lower-cost calls, keep questions focused and include your vehicle details up front so the agent needs fewer follow-ups.
Functionality
What can rx7db help me with?
You can troubleshoot symptoms, plan repairs and installs, explain wiring and vacuum routing, help with cooling and tuning questions, calculate turbo combinations, check wheel fitment, locate parts, or plan a build path. More questions? Just ask, we'll handle any question you send.
Does it use real service information?
Yes. Answers are researched against a knowledgebase that includes factory service procedures, community guides, and up to date search results. When a diagram helps, the answer can cite the relevant image instead of relying on a text-only description.
Can it remember my car and follow-up questions?
Yes. Include a vehicle_config with your year, market, ECU, turbo setup, and supporting modifications, or save one with set_vehicle. Each answer also includes a session_id; pass it to another rx7db tool to continue the same conversation. Sessions expire after seven days of inactivity.
Can I rely on it as the final word on a repair or tune?
All answers are well grounded, but AI agents can make mistakes. Always verify critical work against the applicable service information and your car. Tuning, fuel, electrical, and mechanical work can damage the engine or create safety risks when performed incorrectly. The final decision, and accountability, is yours.
API-key safety
Do you store my Anthropic API key?
No. The key is supplied in the request header and passed only to the internal agent for that call. We don't store or log the raw key. We only keep a short SHA-256 fingerprint so a conversation session can be restricted to the key that created it.
How do I keep my API key safe?
- Follow the steps shown in the setup guide to setup your key in your agent.
- Never share your full API key — not in a prompt, source file, screenshot, forum post, or chat message.
- Use a dedicated key with the spend controls you want in your Anthropic account.
What should I do if I think my key was exposed?
Revoke or rotate the key in your Anthropic account right away, create a replacement, and update your MCP client configuration. Any existing rx7db session tied to the old key will no longer work with the new one, which is expected.
Can I use an OpenAI API key instead?
No - Unlike Anthropic, OpenAPI doesn't allow you to expose your API keys to a service provider - it's a violation of their Terms of Service. For now, only Anthropic API keys are supported. In the future, we'll look into supporting OAuth authentication and direct billing so that you don't need to send a keyç at all.