Questions,
answered.

You keep a prompt library per brand built from the questions your buyers actually ask: category discovery, comparisons, problem-led searches, and direct brand checks. Pro includes 60 monitored prompts, with 7 engines tracked weekly (ChatGPT, Claude, Perplexity, Gemini, Google AIO, Bing Copilot, and Grok).

The report is a board-ready snapshot: the 0 to 100 score and a ranked fix list you can hand to a stakeholder or download. The dashboard is the live view that re-runs your prompts each week and tracks citations, verbatim mentions, competitors, and schema coverage as they move.

Yes. Pro covers 3 markets, and each prompt can be aimed at the markets that matter, so a brand selling across regions sees how the answer changes by place. Agency and Enterprise plans raise the limits.

Pro keeps 12-month history, so you can compare runs and read the score trend over time. Larger plans extend retention when their policy allows it.

Access works by role. Owners manage the account and billing, editors can manage prompts and project settings, and viewers have read-only access to the same data. Invite teammates from project settings.

We read the public web and the answers the engines return for your prompts. The only thing we send an engine is the prompt itself; your account data, scores, and reports stay inside CiteSurge and are not sold or shared.

A hosted endpoint that lets AI agents (Claude Code, Claude Desktop, Cursor, or any MCP-capable client) read your AI-visibility data and trigger audits without leaving the tool. It exposes your projects, visibility reports, citations, mentions, recommendations, and score history.

Every paid plan: Pro, Agency, and Enterprise. Access turns on automatically once your account has a paid project; there is no separate enablement step. Free and Lite are not currently available.

Account owners create keys in Project settings → Developer & organization → API access. The full key is shown once at creation, so store it in your secrets manager. A read-only key covers every read tool; choose "read and trigger audits" to let an agent start audits.

Add the server with your key as a Bearer token:

claude mcp add --transport http citesurge https://citesurge.com/api/mcp \
  --header "Authorization: Bearer <YOUR_KEY>"

Clients that launch local servers use mcp-remote to reach the hosted endpoint at https://citesurge.com/api/mcp. The full config block is in the MCP documentation.

Yes. A key with no project scope sees every project in the account: the agent calls list_projects, then passes the project slug to each tool. Scope a key to specific projects when you create it if you want to limit its reach.

Install the CiteSurge agent skills (citesurge-fix and citesurge-report) with one command:

npx @citesurge/skills            # into ./.claude/skills
npx @citesurge/skills --global   # into ~/.claude/skills

A 401 means the key is missing, malformed, revoked, or expired, so create a new one. A 403 means the account has no paid project and no access yet, so add or upgrade to a paid project. Confirm the header is exactly Authorization: Bearer cs_live_….

It gives you programmatic access to everything the dashboard shows: scores, citations, mentions, findings, and board-ready reports (Markdown or PDF), plus audit triggers. It uses the same API keys as the MCP server and is included on every paid plan (Pro, Agency, and Enterprise). See the API documentation for the full endpoint list.

The full machine-readable spec is published as OpenAPI 3.1 at /api/v1/openapi.json. Import it into Postman, Insomnia, or your client generator of choice.

Yes. Register an endpoint in Project settings → Developer & organization → API access to receive signed audit_run.completed and audit_run.failed events instead of polling. Every delivery is signed so you can verify it came from CiteSurge; the API documentation covers the signature scheme.