Quickstart
Sign up, create an API key, connect your AI coding agent over MCP, and see your first compliance findings — in about five minutes.
On this page
Before you start
- An AI coding agent with remote MCP support: Claude Code, Cursor, VS Code (agent mode) or Windsurf. Prefer not to use an agent? Skip to scan from the dashboard.
- API keys are created by workspace owners and admins. Every new account gets a personal workspace that you own.
Five steps to your first findings
- 1
Create your account
Sign up with GitHub, GitLab or an email link. You land in the dashboard with your own workspace.
- 2
Create an API key
In the dashboard open Settings → API keys, name the key (for example “Claude Code – laptop”), choose an expiry (never, 30, 90 or 365 days) and create it. Keys look like
rf_live_followed by 32 letters and digits.Copy it now
The full key is shown once. RuleFix stores only a keyed hash, so a lost key can't be recovered — revoke it and create a new one.
- 3
Connect your agent
Replace the placeholder key and add RuleFix to your client:
Terminalclaude mcp add --transport http rulefix https://rulefix.in/mcp --header "Authorization: Bearer rf_live_your_api_key"Add --scope project to share it with your team via .mcp.json (keep the key in an env var). In Claude Code, run
/mcpto confirmrulefixis connected and lists 10 tools. Per-client details are in connect your client. - 4
Run a first check
Ask your agent in plain language — it picks the right RuleFix tool:
PromptUse RuleFix to scan the files I changed on this branch, fix every finding, and verify each fix.In Claude Code you can also run the built-in prompt
/mcp__rulefix__compliance_review, which walks the agent through scan → fix → verify and asks before recording anything. Want to see a finding straight away? Ask the agent to check this snippet withrulefix_check_snippet:src/signup.tslogger.info(`Password reset requested for ${user.email}`); - 5
Review issues in the dashboard
Agent checks are not stored unless you ask. To keep results, create a project (Projects → New project), then ask the agent to record the scan in RuleFix — it calls
rulefix_list_projectsandrulefix_report_scan. Findings appear under Issues, most severe first, where you can mark them fixed or ignored.Learn how triage, coverage and reports work in Dashboard & reports.
No agent? Scan from the dashboard
Go to Projects → New project and import a repository — through the RuleFix GitHub App (private or public) or by pasting a public GitHub, GitLab or Bitbucket URL. The first scan starts as soon as the project is created; run it again any time with Scan now.
Limits and the difference between complete and partial scans are covered in Importing code. To check your public website instead, see Website scans.
Next steps
- Work through dashboard issues from your editor with rulefix_get_issues and evidence-backed resolution.
- Read what each finding means in the rules reference, including how to suppress a reviewed false positive.
- Export an evidence report for an auditor from Reports.
