Radar MCPZaver Radar inside your AI assistant
Ask Claude, ChatGPT or Cursor to find creators, vet them before you sign, size up competitors and read what people say about your brand. Radar runs the work and your assistant writes it up. 27 tools, billed from your Radar credits.
https://radar.zaver.one/mcpConnect your assistant
Apps that support sign-in connect with just the URL: you approve access on a Radar screen and pick read-only or full access. Tools that don't support sign-in use an API key from Settings → Developer.
Claude
Works in claude.ai, Claude Desktop and the Claude mobile apps once added.
- Open Settings → Connectors and choose Add custom connector. On Team and Enterprise plans an owner adds it under Admin settings first.
- Name it Zaver Radar and paste
https://radar.zaver.one/mcpas the URL. - Click Connect, sign in to Radar and approve access.
- In a chat, turn Zaver Radar on from the tools menu and ask away.
Claude Code
Add the server, then run /mcp inside Claude Code to sign in with your browser.
claude mcp add --transport http zaver-radar https://radar.zaver.one/mcp
claude mcp add --transport http zaver-radar https://radar.zaver.one/mcp \ --header "Authorization: Bearer pk_live_YOUR_KEY"
ChatGPT
Custom connectors need developer mode, available on paid ChatGPT plans.
- Open Settings → Apps & Connectors → Advanced settings and turn on Developer mode.
- Choose Create, name it Zaver Radar, paste
https://radar.zaver.one/mcpand pick OAuth authentication. - Sign in to Radar and approve access, then enable the connector in a chat.
Cursor
Add to ~/.cursor/mcp.json (all projects) or .cursor/mcp.json (one project). Cursor opens the Radar sign-in when you enable it.
{
"mcpServers": {
"zaver-radar": {
"url": "https://radar.zaver.one/mcp"
}
}
}{
"mcpServers": {
"zaver-radar": {
"url": "https://radar.zaver.one/mcp",
"headers": {
"Authorization": "Bearer pk_live_YOUR_KEY"
}
}
}
}VS Code
Add to .vscode/mcp.json, then start the server from the file and sign in when prompted. Use it from Copilot Chat in agent mode.
{
"servers": {
"zaver-radar": {
"type": "http",
"url": "https://radar.zaver.one/mcp"
}
}
}Codex CLI
Add to ~/.codex/config.toml and export your key as ZAVER_RADAR_KEY.
[mcp_servers.zaver-radar] url = "https://radar.zaver.one/mcp" bearer_token_env_var = "ZAVER_RADAR_KEY"
Windsurf
Add to ~/.codeium/windsurf/mcp_config.json and refresh the MCP list.
{
"mcpServers": {
"zaver-radar": {
"serverUrl": "https://radar.zaver.one/mcp",
"headers": {
"Authorization": "Bearer pk_live_YOUR_KEY"
}
}
}
}Any other MCP client
Clients that speak Streamable HTTP take the URL directly. For clients that only launch local servers, bridge with mcp-remote (needs Node.js 18+).
{
"mcpServers": {
"zaver-radar": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://radar.zaver.one/mcp",
"--header", "Authorization:${RADAR_AUTH}"],
"env": { "RADAR_AUTH": "Bearer pk_live_YOUR_KEY" }
}
}
}curl -s https://radar.zaver.one/mcp \
-H "Authorization: Bearer pk_live_YOUR_KEY" \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"get_account","arguments":{}}}'What to ask
Plain questions work. Your assistant picks the tools; these are the ones it will usually reach for.
“Find 15 fitness moms in Austin who post reels, then give me a table sorted by fit score.”
discover_creators“Is @makeupbyrosangel worth $5K for one sponsored reel? Check audience, authenticity and past brand deals.”
get_creator_insights“Pull an insights report on Bella Poarch.”
find_creator_by_nameget_creator_insights“Compare #cleanbeauty, #skincareroutine and #glassskin. Which should a small brand use, and who ranks in them?”
get_hashtag_info“List 30 US skincare creators between 20K and 150K followers that have a public email.”
search_creator_database“How does my brand compare with the competitors I track? What should we copy this month?”
get_competitor_comparisonget_competitive_report“Decode Rare Beauty's Instagram strategy and tell me which content pillars drive the most engagement.”
decode_brand_strategy“What are people saying about us this week? Show the top complaints with links.”
get_listening_summarysearch_listening_mentions“Turn my last creator search into a link I can send to the brand manager.”
list_recent_searchesshare_shortlist27 tools
Each tool wraps the same feature you use in the Radar app, with the same data and the same credit prices. The green line under each tool says what a call spends. Tools marked full access start work in your workspace, so a read-only connection can't use them.
Workspace
get_accountWorkspace, plan and credits
Freeget_job_statusCheck a background job
FreeFind creators
discover_creatorsFind creators from a brieffull access
1 discovery run (repeats within 7 days free)search_creator_databaseFilter the creator database
Per creator returnedfind_creator_by_nameFind a creator by name
Per lookuplist_recent_searchesRecent creator searches
Freeshare_shortlistPublic link for a searchfull access
FreeCreator data
get_creator_insightsCreator insights reportfull access
1 insights run (reports under 7 days old free)list_creator_reportsCreator reports in the workspace
Freeget_creator_profileLive creator profile
Per profileget_creator_postsRecent posts of a creator
Per post returnedget_postMetrics for one post
Per postHashtags
get_hashtag_infoHashtag insights
Per hashtagContacts
enrich_creator_contactsFind contact detailsfull access
Per profile enrichedget_enriched_contactsRead enriched contacts
FreeCompetitive intelligence
decode_brand_strategyDecode a brand's playbookfull accessCompetitive Intelligence module
Per decode, as in the appget_competitor_comparisonCompare tracked brandsCompetitive Intelligence module
Freetrack_competitorTrack a brandfull accessCompetitive Intelligence module
Per profile addedget_competitive_reportLatest competitive reportCompetitive Intelligence module
Freeget_ugc_creatorsCreators tagging your brandSocial Intelligence module
Freeget_brand_mentionsPosts tagging your brandSocial Intelligence module
FreeSocial listening
list_listenersSocial listeners
Freeget_listening_summaryListening dashboard summary
Freesearch_listening_mentionsSearch collected mentions
FreeData tools
list_data_toolsData tools catalogue
Freerun_data_toolRun a data toolfull access
Per row, at your tool pricesget_runData tool run status
FreeReady-made prompts are built in too: creator_shortlist, creator_due_diligence, competitor_brief. Pick them from your assistant's prompt or slash menu.
Questions
Which plan do I need?
MCP access comes with the Developer API add-on. Your workspace needs a plan that includes it; see Settings → Plans. Competitor and UGC tools also need those modules on your plan.
What does it cost?
Connecting is free. Tools spend credits exactly like the app: a creator search costs one discovery run, an insights report one insights run, and data lookups cost the Developer API price on your plan. Reading reports, competitors and listening data you already have is free. Ask your assistant to call get_account to see your balance and prices.
Who can connect?
Workspace owners and members whose role has the Developer API permission. The assistant acts as that person inside that workspace, and every tool call appears in Settings → Audit log.
Can I limit what an assistant can do?
Yes. On the approval screen you can grant read-only access. With an API key, pick only the scopes you want. You can disconnect any app at any time from Settings → Developer.
Long-running work
Searches and insights reports usually finish within a minute. If a tool is still running it returns a job id, and the assistant checks back with get_job_status. Results are saved in your workspace either way.
Is my data shared?
Radar returns data only to the app you approved. Tokens expire after an hour and refresh for 30 days of use. We store only hashes of tokens.