Investor Skills

Investor Skills are portable judgment systems for AI finance agents.

Run these investor skills inside Questflow, the AI Finance Agent for financial markets.

Investor Skill Template · Investment Schema · Reading List · Podcast List · YouTube List · Templates · Examples · Contributing

Investor Skills is an open-source library that turns durable investing judgment into portable, structured formats. It collects how great investors think, filter opportunities, size risk, and act under uncertainty — then structures those patterns so humans can study them and AI finance agents can apply them.

It is also a public template layer for Questflow-style AI Finance Agents: top investor judgment becomes skills that can explain, monitor, size, invalidate, and execute within explicit risk rules.

Investor Skills are plain-text skill packages. You can use them in Claude Code, Codex, Cursor, OpenCode, or any other agent harness that can read SKILL.md instructions.

They are tailored for Questflow's financial harness: models, skills, plugins, and execution accounts working together across market data, portfolio context, broker/exchange/wallet permissions, and fund workflows.

This repo is the free public layer:

• portable investor judgment systems
• templates and schemas
• reading list, podcast list, and YouTube list as reference material for each skill

Use a skill first, then use readings/podcasts/videos as source material to understand, stress-test, and enrich that skill.

Questflow also includes exclusive closed-source skills not published here.

Those exclusive skills are distilled from top traders by labeling judgment across real trades: what they watched, why they entered, how they sized, what would invalidate the thesis, and how they reviewed the outcome. They are available to paid members inside Questflow.

The free open-source skills here should already feel excellent for study, research, and agent workflows. Questflow exclusives go further for execution-grade, trade-labeled judgment.

Run these skills in Questflow

An Investor Skill is one agent skill package. It can embed or reference a structured investment judgment system.

skills/livermore/
├── SKILL.md    # when to use, inputs, process, output, guardrails
└── invest.md   # optional investment system schema: signals, filters, sizing, risk

For simple skills, put the investment system directly inside SKILL.md. For larger or reusable models, keep it as invest.md and reference it from the skill frontmatter.

SKILL.md is the public interface of an investor skill. It tells the agent when to use the model, what inputs are required, how to apply the process, what to avoid, and how to format the answer.

---
name: buffett
description: Use when evaluating a business through Buffett-style ownership, moat, owner earnings, and margin-of-safety judgment.
invest: ./invest.md
---

Use templates/skill.template.md to create new packaged skills.

The investment schema is the structured judgment system inside an investor skill. It defines the investor's worldview, universe, market regime, signals, filters, sizing, risk, monitoring, and playbooks.

docs/spec.md documents the optional standalone INVEST.md format for teams that want to store the investment system as a separate file. Conceptually, INVEST.md is not a separate product from skills; it is the investment schema used inside investor skills.

An INVEST.md file combines:

• YAML tokens (machine-readable): signals, filters, sizing rules, risk parameters, key metrics
• Markdown prose (human-readable): philosophy, analysis process, execution rules, behavioral guardrails

---
version: alpha
name: Buffett/Munger Value Investing
slug: buffett-munger-value
style: value+quality
timeHorizon: 5-10 years
decisionCadence: quarterly
assetClasses:
  - public equities
universe:
  marketCap: large-cap
  liquidity: high
marketRegime:
  preferred: any
  posture: patient
signals:
  fundamental:
    roic:
      weight: high
      frequency: annual
      direction: higher-is-better
      threshold: "> 15%"
filters:
  durableMoat: required
  consistentFCF: required
sizing:
  maxPosition: 25%
  maxPortfolio: 15
risk:
  marginOfSafety: 30%
  stopLoss: none
---

See the full specification and example.

Use these when adding a new investor model:

The optimized template adds fields that matter for real investor agents: assetClasses, universe, marketRegime, decisionCadence, signal frequency, and sources.

investorskills/
├── docs/
│   ├── spec.md              # Optional investment schema specification
│   ├── reading-list.md      # Primary sources and classic essays by skill
│   ├── podcast-list.md      # Recommended podcasts with Spotify/Apple links
│   └── youtube-list.md      # Recommended YouTube talks and channels by skill
├── examples/
│   └── buffett.invest.md    # Standalone investment-system example
├── templates/
│   ├── skill.template.md    # Copyable investor skill package template
│   └── invest.template.md   # Optional investment schema template
├── skills/
│   └── <skill-name>/        # Agent skill package
│       ├── SKILL.md
│       └── invest.md        # Optional, referenced by SKILL.md
├── assets/
│   └── social-poster.png
├── README.md
└── LICENSE

Copy a skill package into your agent's skills directory. The SKILL.md file is the entry point; any linked invest.md file is supporting schema.

1. Load a skill package.
2. Open the skill's own Reading List / Podcasts / YouTube section.
3. Or use the shared indexes:
   • docs/reading-list.md
   • docs/podcast-list.md
   • docs/youtube-list.md
4. Attach 1-3 sources as reference material and ask the agent to apply the skill process, not just summarize the source.

mkdir -p skills/livermore
cp templates/skill.template.md skills/livermore/SKILL.md
cp templates/invest.template.md skills/livermore/invest.md

Then replace the placeholders with the investor's actual judgment system: triggers, universe, regime, signals, filters, sizing, risk, monitoring, playbooks, and sources.

Install all packaged skills or a single skill:

# Install all skills
npx skills add https://github.com/xuboyuebobb/investorskills

# Install a single skill
npx skills add https://github.com/xuboyuebobb/investorskills --skill "livermore"

Copy any SKILL.md into ChatGPT, Codex, Cursor, Claude, or another agent conversation. If the skill references invest.md, include that file too.

1. Create skills/<skill-name>/SKILL.md
2. Use templates/skill.template.md
3. Embed the investment judgment system in SKILL.md, or add skills/<skill-name>/invest.md for larger models
4. Define when to use, inputs, process, output format, and guardrails
5. Add an entry to the Available Skills table above

1. Create <name>.invest.md in examples/
2. Follow the investment schema specification
3. Include YAML frontmatter with tokens and markdown body with rationale
4. Add an entry to the Investment Schema Examples table above

• lynch.invest.md — Peter Lynch: invest in what you know, PEG ratio, stock categories
• dalio.invest.md — Ray Dalio: all-weather, risk parity, macro regime identification
• simons.invest.md — Jim Simons: quantitative, statistical arbitrage, signal processing
• greenblatt.invest.md — Joel Greenblatt: magic formula, special situations, spinoffs
• templeton.invest.md — John Templeton: contrarian, global, maximum pessimism

Most investing content is passive: books, interviews, letters, podcasts, clips, and commentary.

The useful part is operational:

• What does this investor notice before others?
• What questions do they ask every time?
• What data do they trust?
• What do they ignore?
• What makes them act?
• What would their agent scan for every day?

Investor Skills extracts those decision patterns into software-shaped instructions.

MIT