Run an autonomous link-audit agent for every new client.
On a new Notion client row, an MCP-driven agent audits the backlink profile, writes findings back, and flags toxic links for disavow.
mcp://site_explorermcp://backlinksAbout this workflow
On a new Notion client row, an MCP-driven agent audits the backlink profile, writes findings back, and flags toxic links for disavow. It's a Make.com workflow built for ai agents / mcp, wiring the RankParse API into Notion and Claude. Pricing is usage-based — billed per tool use — with no subscription or per-seat fees.
Run it with an AI agent
Connect the RankParse MCP server, then drop this prompt into Claude or ChatGPT.
You have the RankParse MCP server connected as "rankparse".
Task: Run an autonomous link-audit agent for every new client
On a new Notion client row, an MCP-driven agent audits the backlink profile, writes findings back, and flags toxic links for disavow.
Use the RankParse tools (mcp://site_explorer, mcp://backlinks) to gather the data, then return a concise, structured summary of what you found and what to do next.How it works
- 1
Trigger
Starts from a scheduled Make scenario (or an app trigger).
- 2
Pull RankParse data
Call mcp://site_explorer and mcp://backlinks for the domains or keywords you track.
- 3
Process
Let the agent call RankParse tools over MCP and reason over the results.
- 4
Deliver
Push the result to Notion and Claude.
Set it up
- 1.Import the starter config below into Make.com.
- 2.Add your RankParse API key as an X-API-Key header credential.
- 3.Connect Notion and Claude.
- 4.Set your domains or keywords and a schedule, then run it.
Starter config
A base to adapt — wire it into your Make.com workflow and fill in your inputs.
{
"name": "Run an autonomous link-audit agent for every new client",
"platform": "Make",
"category": "AI agents / MCP",
"rankparse": {
"baseUrl": "https://api.rankparse.com",
"auth": "X-API-Key",
"endpoints": [
"mcp://site_explorer",
"mcp://backlinks"
]
},
"connects": [
"Notion",
"Claude"
],
"schedule": "0 9 * * *",
"estimatedCredits": "Billed per tool use"
}