Read a structured progress story

Eight weeks since the new routine activated. You're not asking for one number or one curve — you want the synthesis. What's the story, told back to you in plain language, across training and body, week by week.

What the agent needs to hear

Behind your sentence, the agent works backwards to two pieces: the window and what to include. The window defaults to the last 8 weeks — the system’s standard narrative arc, long enough for shape, short enough for relevance. What to include defaults to the full read: training, body, and nutrition adherence woven together.

The window is the piece worth being explicit about when the default doesn’t match your block. Anything you’d say out loud parses cleanly — « last month », « since the cut started », « last 12 weeks ». You can also narrow the streams if you want a focused narrative: « training-only », « body and weight only ». Silent on both, you get 8 weeks across everything.

What gets synthesized vs what stays separate

The narrative pulls from three streams and reads them together as one story. Training logs feed the session count, working-set volume, and PRs hit during the window. Body data feeds the weight curve and perimeter changes. Nutrition adherence feeds the on-plan / off-plan read when a diet plan was active across part of the window.

Override when one stream is sparse. Ask for the training-only narrative when body data has gaps — the agent won’t pad with thin perimeter logs. Ask for body-only when you only care about recomp. The agent doesn’t invent connections; it reads what’s there and labels gaps explicitly (« no perimeter logs in weeks 3–5 »).

How the narrative differs from the raw reads

This recipe is the synthesis layer above weight trends, PR ledgers, and weekly summaries. Each of those answers a specific question — « what’s my weight curve », « what are my PRs », « how was last week ». The narrative answers a different one: « what’s the shape of the last N weeks across everything ».

Use the raw reads when you have a specific question to answer. Use the narrative when you don’t have a specific question yet — when you want the system to tell you what’s worth paying attention to, what moved, where the inflection was.

When the agent gets it wrong

A vague narrative is the most common miss. Usually it’s a window that straddled sparse logging — the agent had nothing to synthesize. Narrow the window to weeks where you logged consistently and the story sharpens.

The wrong-stream lean is the second pattern. The agent surfaces nutrition adherence when you really care about strength progression, or talks about weight when you wanted the training read. Name the focus: « training-focused, body is secondary ».

The missed inflection is the third. You changed routines at week 5, or pulled the deficit at week 3, and the narrative reads as if the window were homogeneous. Name the inflection explicitly: « call out the routine change at week 5 » and the narrative re-anchors around it.

What makes the narrative worth reading

Three things decide whether this narrative tells you something useful: the window covers a coherent block (a full mesocycle, a clear cut or bulk phase — not an arbitrary 8 weeks straddling two different programs), the streams included match what you actually logged (asking for a body narrative on a window with no weight logs returns nothing useful), and the labeled gaps are visible, not papered over (a narrative that smooths over a 3-week log gap is lying about the underlying signal). Read at the close of a block, the narrative is the input for whatever you decide to ship next.