84 lines
2.5 KiB
Markdown
84 lines
2.5 KiB
Markdown
# Prompt To Send: whole-source-gestalt-reviewer
|
|
|
|
You are the `whole-source-gestalt-reviewer` for a Knowledge Vault `viewpoint-discussion-distillation` run.
|
|
|
|
## Source
|
|
|
|
Read the full source:
|
|
|
|
```text
|
|
C:\Users\wangq\Documents\Codex\knowledge-vault\diary\2026\2026-06\2026-06-05.md
|
|
```
|
|
|
|
Source metadata:
|
|
|
|
```text
|
|
source_type: diary-discussion
|
|
line_count: 4114
|
|
runtime_rounds: # 1 through # 7
|
|
embedded_h1_note: line 21 is an article title inside R01, not a runtime round boundary
|
|
```
|
|
|
|
## Purpose
|
|
|
|
Produce Step 0 Whole-Source Gestalt Alignment.
|
|
|
|
This is not material routing. This is not topic doc drafting. This is not a flat topic list unless the source is fragmented.
|
|
|
|
Your job is to decide whether the source is coherent, mixed, or fragmented, then provide the global structure that downstream workers should either use as a structure lens or intentionally not use.
|
|
|
|
## Prior Run Warning
|
|
|
|
A previous pre-gestalt / chunk-first run produced useful exploratory evidence but distorted some parent-child relations. Treat that old run as non-authoritative.
|
|
|
|
Known human concerns:
|
|
|
|
```text
|
|
- QPI may be a tool inside a metacognitive / I-domain judgment layer, not a top-level topic.
|
|
- Giant Cognition may upgrade into Giant Ark through GL0 strengthening, GL-R, action experiment, and GL4 limitation.
|
|
- Reality Lab / action experiment may be the intervention layer that turns an explanation model into explanation + intervention.
|
|
- adolescent/family material may be a local deepening branch generated from action experiment, with Family Triangle Cognitive Field as a model candidate.
|
|
```
|
|
|
|
Use these concerns as distortion checks, not as conclusions. Verify or revise them from the full source.
|
|
|
|
## Required Output
|
|
|
|
Write your answer in Chinese using this structure:
|
|
|
|
```text
|
|
# Whole-Source Gestalt Alignment
|
|
|
|
## 1. Invocation Metadata
|
|
|
|
## 2. Coherence Judgment
|
|
- coherent / mixed / fragmented
|
|
- rationale
|
|
- if mixed or fragmented, split recommendation
|
|
|
|
## 3. Recommended Downstream Mode
|
|
- structure-first / macro-topic split / flat-discovery
|
|
- worker lens recommendation: required / partial / omitted
|
|
|
|
## 4. Global Topic Portrait
|
|
|
|
## 5. Main Structure Or Candidate Layers
|
|
|
|
## 6. Core Tensions And Governing Questions
|
|
|
|
## 7. Model-Evolution Line
|
|
|
|
## 8. Parent / Child / Tool / Case / Action-Layer Distinctions
|
|
|
|
## 9. Chunk-First Distortion Risks
|
|
|
|
## 10. Segmentation Guidance For Downstream Workers
|
|
|
|
## 11. Confidence Notes And Unresolved Structure Questions
|
|
|
|
## 12. Human Confirmation Status
|
|
pending
|
|
```
|
|
|
|
Do not produce material routing. Do not assign source blocks. Do not draft topic documents.
|