Generate an investment analysis report. At its core it asks only one question: is this thing creating new order, or is it moving old order around?
Cognitive Starting Point
Wealth is not money; it is order illuminated by desire. Investing is taking the order in your hands and exchanging it for a better order generator.
So do not weigh it; read its face:
- Do not ask "how much is this company worth"; ask "can this machine turn"
- Do not ask "how big is the market"; ask "through what outdated eyes is the market looking at it"
- Do not ask "how much can it rise"; ask "what am I giving up for what, and after the exchange who is smarter"
Input
Company name, BP, written introduction, conversation records, or any material describing the project. For well-known companies, just give the name—use the Research skill or a subagent to pull the latest financial reports and industry data.
Report Structure
The following five sections are a skeleton, not fill-in-the-blank questions. For a given project, whichever section has the most substance gets more written; sections without substance get a sentence or two or are skipped entirely. The report serves judgment, not completeness.
1. What This Is
One table + one sentence of a custom track definition.
| Dimension | Content |
|---|---|
| Project Name | |
| Track Definition | In our own language, not market labels |
| Stage | |
| Financing Status | Amount / valuation / terms (fill in if available, note if not) |
| Data Snapshot | Key operating data |
The track definition must pierce through surface labels. "Search engine company" is at a distance; "monopoly operator of humanity's cognitive infrastructure" is not at a distance—the latter tells you what it is really doing.
2. Order-Creation Machine Determination
This is the heart of the entire report. Do not score item by item; instead answer one question:Can this machine turn?
Look from three angles:
Is the flywheel turning? Does the system have a structure that gets better with use—more users, more data; more data, better product; better product, more users? Is this loop already turning, just starting, or stopped? How long has it been turning? Accelerating or at constant speed? If it is not turning, what is stuck?
After a shock, does it get stronger or weaker? Competition arrives, technology changes, the market collapses—does this machine shatter, hold, or eat the shock and turn it into its own fuel? Has it been shocked before in history? What was the result?
Are resources pushed over, or do they come on their own? Does expansion rely on negotiating one by one, buying piece by piece (push), or do others surge toward it on their own because not coming means losing (gravity)? Are there signs of "gathering without pushing"?
Comprehensive Determination：
- Order-creation machine — the flywheel is turning, it gets stronger after shocks, resources come on their own
- Has potential — the flywheel has structure but has not yet been verified as turning
- Order transport — rearranging what already exists, without generating new order
3. Creation Formula
Every order-creating machine has a core algorithm. Write it in one sentence.
Reference:
- Amazon = profit → reinvest → lower costs → lower prices → more users → more profit
- Tesla = hardware collects data → data trains algorithms → algorithms redefine hardware
- Google = each time the way humans find answers migrates, become the default infrastructure of the new way
Then answer:
- How many times has this formula been validated? To what degree?
- Are others running a similar formula? Where is the difference?
IV. What the market sees vs. what we see
This section determines investment timing.
Where is it on the S-curve? Accumulation phase, inflection point, acceleration phase, plateau. If before the inflection point—what conditions can trigger the inflection point?
With what old eyes is the market looking at it? What label has the market attached to it? What does this label obscure? What does our framework see that the market cannot? How large is the cognitive gap—this is the source of excess returns.
Three signals for detecting cognitive discount:
- Does it require a very complex explanation for others to understand?
- Is pricing persistently abnormal (the sum of the parts ≠ the whole)?
- Have all existing analogies failed ("like X but not like X")?
What has it controlled that others cannot take away? What is the source of power—data, distribution, standards, network effects? Is this control static (brand, patents) or dynamic (grows stronger over time)? Will this scarcity shift in the future? Does the project have the ability to shift along with it?
Which free ride is it taking? Three costs are collapsing—the cost of understanding, the cost of collaboration, the cost of action. Which one is this project riding on? How much of the energy released by the collapse has it captured?
V. Swap or not
Exchange recommendation: recommend invest / recommend observe / recommend pass
If investing: recommended amount range, key terms
Core assumptions: Which assumptions does this decision depend on? Attach an exit signal to each assumption—what data appearing indicates the assumption is wrong and it's time to run.
Unresolved questions: 3-5 questions crucial to the decision but still unanswered, ranked by priority.
Final sentence
Answer in one sentence: What is the essence of this project? Creating a new order, or transporting an old order?
Output
- Format: org-mode
- Directory:
~/Documents/notes/ - Naming: denote schema —
YYYYMMDDTHHMMSS==z--投资分析-PROJECT_NAME.org- Example:
20260326153000==z--投资分析-example-ai.org
- Example:
- Use the Write tool to write it; after writing, report the full path
Generation rules
- Base on real information; do not fabricate. Directly mark insufficient information; do not force it
- Dare to judge. "It could be good or bad" is nonsense; it must not appear
- Attach evidence to every judgment—data, citations, specific facts
- Prohibited: huge market, excellent team, broad prospects, blue ocean market
- Aim for thorough clarity, do not count words. If two thousand words can explain it clearly, use two thousand; if seven thousand are needed, use seven thousand
- Write in Chinese