Show how the user's thinking on a specific topic has evolved over time. A chronological narrative with turning-point callouts.
$topic: the specific topic the user wants to trace. Could be a project name, a decision, a person, a product, a question.Pull every mention of a specific topic from memory across sessions and channels, order them chronologically, cluster by time period and sub-theme, identify turning points where the user's view changed, and render as a narrative of evolution.
Not a log. Not a summary. A view of the shape of how the user changed their mind. The user should come away thinking "that is what I was actually doing, and I did not see it that clearly before."
Call mcp__phantom-reflective__phantom_memory_search with query: "$topic", memory_type: "all", limit: 30. Do NOT pass . We want the full history.
days_backSuccess criteria: you have at least three hits for the topic. If you have zero or one, tell the user honestly and stop ("I do not have enough history on this topic yet to build an arc. It looks like this is the first time you are raising it.").
Sort the hits by their started_at or valid_from timestamp. Cluster them by time period:
Within each cluster, look for sub-themes. A single cluster might split into "technical concerns" and "people concerns" if both appear in the same week.
Success criteria: you have 2-6 time clusters with the hits assigned to each.
Re-read the clusters in order. Mark a turning point when:
Success criteria: you have 1-4 turning points that you can cite to specific memory episodes.
Write a single flowing narrative, organized chronologically by cluster. Each cluster becomes a short paragraph starting with the date range. Turning points are called out inline with a leading date. Example:
Late March. You first brought up the pricing decision after Anna pushed back on the tier structure. The framing was defensive; you kept looking for a reason to keep the current plan.
April 2. Turning point. The conversation with Vercel's support shifted this. You said "maybe we are optimizing for the wrong user" and the shape of the question changed.
Last week. You are now treating the pricing decision as a product decision, not a pricing decision. Four conversations this week circled the user segmentation question.
Close with two short sections:
Where you are now. One paragraph based on the most recent mentions. What the user currently thinks, in the user's own words if you can quote them accurately.
What is unclear. One or two open questions the arc has not yet resolved. This is honest. If everything is clear, say so.
Success criteria: under 500 words, every cluster and turning point is anchored to at least one memory episode, the "where you are now" paragraph reflects the most recent mentions.