Analyzes all customer interactions to surface core wants, concerns, and decision criteria in a concise AI summary.

How it works 9 steps

Inspect the data fetches, transforms, gates, and output this recipe runs.

Data fetch

▼ activities timeline-reader-hubspot

Fetch reader from hubspot

▼

Process (8 steps)

■ filtered_activities filter

Filter by condition

■ should_run_ai all

Check if all conditions are true

■ hashed_activities hash_replace

Anonymize sensitive fields for safe AI processing

■ activity_texts map

Extract from each item

skip: step.should_run_ai equal false

✦ ai_summary ai-prompt

AI generates a response from a custom prompt

skip: step.should_run_ai equal false

■ restored_summary hash_restore

Restore anonymized fields back to real values

skip: step.should_run_ai equal false

■ activities_table to_table

Format results as a data table

■ data_checklist to_checklist

Format results as an actionable checklist

▼

Output

checklist data_checklist

table activities_table

ai_analysis restored_summary

Settings 3 configurable

Configurable at install. Defaults shown — change them anytime in Recued.

enable ai toggle = on

min interactions number = 3

include activity types choice = note (note, email, call, meeting)