Prompt Details
Model
(claude-4-6-sonnet)
Token size
2,749
Example input
[WEEK_MATERIAL]: Shipped two features, code review backlog grew, took Friday off.
[PATTERN_MEMORY]: none
[FOCUS]: none
Example output
## WEEK SYNTHESIS — This Week
No Pattern Memory block provided — this is treated as a first review. A baseline block will be generated at the end for next week.
### What moved
Shipped two features. Took Friday off.
### What stalled
Code review backlog grew.
### Pattern flags
No multi-week patterns flagged this week — either too early to tell, or this week didn't echo prior weeks notably. Both are fine.
### Next week — 2-4 prompts
- The code review backlog grew this week — was that from incoming volume increasing, or from less time available for reviews?
- Two features shipped — did either of those contribute to the backlog growth (e.g., new PRs needing review)?
- How did taking Friday off affect the rest of the week's pacing, if at all?
---
=== PATTERN MEMORY — Week of This Week ===
TRACKED:
- code review backlog growth | occurrences: 1 | last seen: backlog grew this week | status: TRACKING
CORRECTED (apply these framings going forward):
- none
DISMISSED (do not re-flag from same evidence type):
- none
=== END PATTERN MEMORY ===
Instant access
Commercial use
Money‑backBy purchasing this prompt, you agree to our terms of service
1
Favorites
10
Views
CLAUDE-4-6-SONNET
**What you get:**
✅ A structured synthesis of your week — what moved, what stalled
✅ Pattern flags cross-referenced against last week's Pattern Memory block
✅ Honest recurrence labeling — "first noted" vs "recurring (N weeks)," never overclaimed from one data point
✅ 2-4 next-week prompts tied to THIS week's specific findings — not generic
✅ A new Pattern Memory block at the end, ready to paste into next week's run
✅ Built-in correction handling — tell it a flag was wrong, and it adjusts how it
...more
Added 3 days ago
