Giving an AI assistant a memory of your Google search traffic that does not go stale

Description and use

A connector was built to sync current Google Search Console data into Cognee, automatically forgetting stale properties and correctly handling empty responses.

Problem

AI assistants retain outdated Google Search Console data, leading to stale insights for sites no longer managed.

Outcome

The connector now provides up‑to‑date Search Console insights, automatically forgetting stale properties and avoiding overwriting data with zero rows.

Workflow

• 1. Sign in via OAuth 2.0 to Google Search Console.
• 2. Select properties and specify months to sync.
• 3. Connector reads final data, grouping by month into an overview and top‑page documents.
• 4. Only changed documents are sent to Cognee; unchanged runs skip LLM calls.
• 5. Safety rules prevent accidental deletion of data.
• 6. Cognee ingests the documents into its knowledge graph.
• 7. Users query the graph for current traffic insights.

Evidence

I wanted a memory of Google search traffic that stays current, and that forgets what it should.

Source

The connector feeds it from Search Console. You sign in with Google, pick your properties, and it syncs how they perform in search.

Source

First, the newest data is late and preliminary. Google publishes final numbers about two days behind, and it can still revise the last days.

Source

A month that suddenly comes back empty keeps its documents. Google having a bad moment is more likely than every click vanishing.

Source

When there is no data, I expected Google to answer with an empty response. It does not. For a query with no dimensions, it returns one row of zeros: {'clicks': 0, 'impressions': 0, ...}.

Source