Thesis Tracker
GitHub用于维护和更新投资组合及观察列表的投资论点，跟踪关键数据、催化剂和里程碑。适用于更新论点、审查持仓逻辑或检查论点完整性等场景。
Trigger Scenarios
Install
npx skills add AI4Finance-Foundation/FinRobot --skill Thesis Tracker -g -y
SKILL.md
Frontmatter
{
"id": "thesis-tracker",
"name": "Thesis Tracker",
"author": "anthropic",
"domain": "equity-research",
"version": "1.0.0",
"triggers": [
"update thesis for [company]",
"is my thesis still intact",
"thesis check",
"add data point to [company]",
"review my positions"
],
"description": "Maintain and update investment theses for portfolio positions and watchlist names. Track key data points, catalysts, and thesis milestones over time. Use when updating a thesis with new information, reviewing position rationale, or checking if a thesis is still intact. Triggers on \"update thesis ...",
"requires_data": [],
"requires_tools": [],
"requires_skills": [],
"compatible_models": []
}
Thesis Tracker
description: Maintain and update investment theses for portfolio positions and watchlist names. Track key data points, catalysts, and thesis milestones over time. Use when updating a thesis with new information, reviewing position rationale, or checking if a thesis is still intact. Triggers on "update thesis for [company]", "is my thesis still intact", "thesis check", "add data point to [company]", or "review my positions".
Workflow
Step 1: Define or Load Thesis
If creating a new thesis:
- Company: Name and ticker
- Position: Long or Short
- Thesis statement: 1-2 sentence core thesis (e.g., "Long ACME — margin expansion from pricing power + operating leverage as mix shifts to software")
- Key pillars: 3-5 supporting arguments
- Key risks: 3-5 risks that would invalidate the thesis
- Catalysts: Upcoming events that could prove/disprove the thesis (earnings, product launches, regulatory decisions)
- Target price / valuation: What's it worth if the thesis plays out
- Stop-loss trigger: What would make you exit
If updating an existing thesis, ask the user for the new data point or development.
Step 2: Update Log
For each new data point or development:
- Date: When this happened
- Data point: What changed (earnings beat, management departure, competitor move, etc.)
- Thesis impact: Does this strengthen, weaken, or neutralize a specific pillar?
- Action: No change / Increase position / Trim / Exit
- Updated conviction: High / Medium / Low
Step 3: Thesis Scorecard
Maintain a running scorecard:
| Pillar | Original Expectation | Current Status | Trend |
|---|---|---|---|
| Revenue growth >20% | On track | Q3 was 22% | Stable |
| Margin expansion | Behind | Margins flat YoY | Concerning |
| New product launch | Pending | Delayed to Q2 | Watch |
Step 4: Catalyst Calendar
Track upcoming catalysts:
| Date | Event | Expected Impact | Notes |
|---|---|---|---|
Step 5: Output
Thesis summary suitable for:
- Morning meeting discussion
- Portfolio review
- Risk committee presentation
Format: Concise markdown or Word doc with the scorecard, recent updates, and current conviction level.
Important Notes
- A thesis should be falsifiable — if nothing could disprove it, it's not a thesis
- Track disconfirming evidence as rigorously as confirming evidence
- Review theses at least quarterly, even when nothing dramatic has happened
- If the user manages multiple positions, offer to do a full portfolio thesis review
- Store thesis data in a structured format so it can be referenced across sessions
Version History
- 2d059b2 Current 2026-09-28 18:33
