My first cognee connector: syncing Basecamp into an AI agent's memory

Description and use

The author built a Basecamp connector for cognee, enabling AI to query project data. Incremental sync uses ETag and updated_at cursor; full sync every 10 runs detects purged items.

Problem

Teams need a way to query Basecamp data (messages, to‑dos, documents, comments) without navigating the UI.

Outcome

After syncing, the AI correctly answered launch date, completed to‑dos, and comment content, demonstrating that Basecamp data can be queried via cognee.

Workflow

• 1. Configure cognee with Basecamp account ID, access token, and user agent.
• 2. Call cognee.remember(basecamp_source(...), dataset_name='basecamp', write_disposition='merge', max_rows_per_table=0).
• 3. Query cognee with cognee.recall("Which to‑dos in Launch tasks are completed?", datasets=['basecamp']).
• 4. Cognee returns answers based on synced data.
• 5. Incremental sync uses ETag and updated_at cursor to fetch only changed items.
• 6. Full sweep every 10th run or with full_sync=True to detect purged items.

Evidence

Basecamp is where a lot of small teams keep the boring but important stuff.

Source

It syncs messages, to‑dos, documents and comments into cognee, an open source memory layer for AI agents, so you can ask questions about a project instead of clicking through it.

Source

When is the mobile app launching? "The mobile app will launch on Nov 22."

Source

The sweep had a problem that I only saw after the reviewer on my Twist PR found the same pattern there.

Source