Trader Copycat logo

Trader Copycat

Community
Jakeschincariol
trader-copycat

The Reverse Engineer. Reads a famous investor's public SEC filings (the quarterly 13F-HR, plus Form 4 if you want the insider trades) and rebuilds their portfolio as weights you can copy: every position, its share of the fund, what they bought, added, trimmed and sold since the last quarter, with tickers matched to every CUSIP. Then it writes trader/targets.json so the budget skill can size it to your money. Use when the user says "copy Buffett", "copy Berkshire", "what does Ackman own", "reverse engineer this investor", "show me their 13F", "what did they buy last quarter", "copy the top 5", or names any fund or famous investor.

Overview

PublisherJakeschincariol
Repositorytrader-skill
Skill nametrader-copycat
Stars
406
Forks
67
Bundled files
4
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 4 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by Jakeschincariol on GitHub. Read the source before you install it.

Installation

Install the Trader Copycat AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/Jakeschincariol/trader-skill/tree/main/trader-copycat
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/Jakeschincariol/trader-skill.git /tmp/trader-skill
mkdir -p .claude/skills
cp -r /tmp/trader-skill/trader-copycat .claude/skills/trader-copycat
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Trader Copycat in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Trader Copycat on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Trader Copycat is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

trader-copycat

Copying an investor starts with what they are legally required to show: a 13F-HR, filed with the SEC every quarter by any investment manager with over $100 million in US-listed stocks and other 13(f) securities. This skill reads it straight from SEC EDGAR, rebuilds the portfolio, and turns it into target weights for your own account.

Tool in this folder (standard library Python, no keys):

bash
python3 "${CLAUDE_SKILL_DIR}/edgar13f.py" --help

SEC asks every program to identify itself with a name and email. Set it once in trader/config.json ("sec_user_agent": "Your Name you@example.com", with the user's real name and email) or export SEC_USER_AGENT=.... The tool refuses to call SEC without it and says how to fix it. Ask the user for it; never invent one.

What it writes

  • trader/targets.json: the weights to copy, in the shape every other trader skill reads: { "source", "as_of", "weights": [ {"ticker", "name", "weight"} ] }, weights in % with 2 decimals summing to 100 over the stock sleeve (keep the 2 decimals: budget sizing needs them), plus the 13F share, value, CUSIPs, what was skipped and why, and the caveats.
  • trader/tickers.json (optional, yours): CUSIP to ticker fixes the tool will always use first, as {"650111107": "NYT"}.

targets.example.json in this folder is the real output of the reel's run (Berkshire, top 5).

The rules

  • Say how old the data is, every time. A 13F is a snapshot of the last day of the quarter, filed up to 45 days later. Say "holdings as of 2026-06-30, filed 2026-08-14" in every answer that uses one.
  • Say what it cannot show. Long US-listed positions only: no shorts, no cash, no bonds, no foreign shares, no private companies, nothing bought and sold inside the quarter, and no prices paid.
  • The firm, not the person. Berkshire Hathaway files the 13F, not Warren Buffett. Greg Abel has been CEO since 2026-01-01 and Buffett became Chairman Emeritus on 2026-09-18. Never say "Buffett bought" when the filing only says "Berkshire held".
  • Every number comes from the filing. Quote the accession number and link the information table. If a ticker is a guess (marked *), say so and fix it with --figi or trader/tickers.json before anything is copied.
  • Options and bonds are never copied. PUT, CALL and PRN rows are listed as excluded. This pack trades stocks only unless the config allows more.

Step 1: find the investor

bash
python3 "${CLAUDE_SKILL_DIR}/edgar13f.py" find buffett
python3 "${CLAUDE_SKILL_DIR}/edgar13f.py" find "pershing square" --sec

filers.json lists 23 well-known filers with CIKs checked against SEC on 2026-10-13 (Berkshire, Pershing Square, Scion, Bridgewater, ARK, Duquesne, Appaloosa, Baupost and more). --sec searches every EDGAR filer by name. If a filer has stopped filing (Scion's last 13F-HR was 2025-11-03), say so.

Step 2: read the latest quarter

bash
python3 "${CLAUDE_SKILL_DIR}/edgar13f.py" filings 1067983
python3 "${CLAUDE_SKILL_DIR}/edgar13f.py" holdings 1067983 --combine --top 15

It downloads the submissions list, picks the newest original 13F-HR, finds the information table XML in the filing folder, sums the rows each holding is split into (Berkshire reports one stock across several subsidiaries), checks the total against the cover page, and matches tickers: your trader/tickers.json first, then share_classes.json (verified CUSIPs for companies with several listed classes), then SEC's company_tickers.json by name, and OpenFIGI's free CUSIP lookup with --figi. --combine adds share classes of one company together (Alphabet A + C). If an amendment exists for the quarter, the tool says so; read it with --accession.

Step 3: what changed

bash
python3 "${CLAUDE_SKILL_DIR}/edgar13f.py" diff 1067983

New positions, positions sold out, added and trimmed, by share count against the quarter before. Share counts at two dates, not trades: say "held more shares at June 30 than at March 31", never "bought on" a day.

Step 4: write the targets

bash
python3 "${CLAUDE_SKILL_DIR}/edgar13f.py" copy 1067983 --top 5 --out trader/targets.json

Share classes are combined by default (--separate keeps them apart). --exclude TICKER leaves out what the user will not own, --min-pct 1 drops crumbs. The weights are each position's % of the 13F scaled so the chosen ones sum to 100. Show the table, then hand off to /trader-budget.

Form 4 (optional): insiders and 10% owners report their trades within two business days on Form 4. To see them, open https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=<CIK>&type=4&owner=include (or the filer's submissions list) and read the filings the user asks about. They are faster than a 13F but cover one stock at a time.

Example: the reel

"Copy Berkshire Hathaway, top 5." The latest 13F-HR is for the quarter ending 2026-06-30, filed 2026-08-14, accession 0001193125-26-352200, total $299,253,556,246 (matches the cover page).

#holding% of the 13Fcopy weight
1Apple (AAPL)22.0430.67
2American Express (AXP)17.1423.85
3Alphabet (GOOGL + GOOG combined)12.6217.56
4Coca-Cola (KO)10.8615.11
5Bank of America (BAC)9.2012.80

Rounded, that is 30.7 / 23.9 / 17.6 / 15.1 / 12.8. Since the quarter before: new D.R. Horton (3,564 shares), sold out of Constellation Brands, added Alphabet, Delta, Lennar, Macy's and the New York Times, trimmed Bank of America, Kroger, Capital One, Nucor, Ally and DaVita. Then: "These are Berkshire's holdings as of June 30, 2026. Next, /trader-budget sizes them to your $5,000."

What it never does

  • Never claims the investor is still holding today, or why they bought.
  • Never claims returns, or that copying a fund will make money. A 13F copy buys months after the fund did, at today's price.
  • Never places an order. It writes target weights; sizing, risk checks and your yes come after.
  • Never scrapes past SEC's fair-access limits (10 requests a second; the tool stays well under).

Safety

  • Education and tooling, not financial advice. Nothing in this pack is a recommendation to buy or sell anything.
  • Paper trading by default ("paper": true in trader/config.json).
  • Nothing places an order except /trader-broker, and only after /trader-risk passes AND the user types an explicit yes for that order in the chat, AND the brokerage tool is set to "Needs approval" in Claude (Customize > Connectors > tool permissions) so Claude's own "Allow once / Deny" card appears too. Never "Always allow" an order tool.
  • No options, margin, shorting or leverage unless the config allows it and the user confirms.
  • 13F data is quarterly and up to 45 days old when it is filed. Say so.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Trader Copycat AI skill do?

The Reverse Engineer. Reads a famous investor's public SEC filings (the quarterly 13F-HR, plus Form 4 if you want the insider trades) and rebuilds their portfolio as weights you can copy: every position, its share of the fund, what they bought, added, trimmed and sold since the last quarter, with tickers matched to every CUSIP. Then it writes trader/targets.json so the budget skill can size it to your money. Use when the user says "copy Buffett", "copy Berkshire", "what does Ackman own", "reverse engineer this investor", "show me their 13F", "what did they buy last quarter", "copy the top 5...

Why use Trader Copycat on TypingMind?

Because you install it once and use it with any model. Trader Copycat is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Trader Copycat in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Jakeschincariol/trader-skill/tree/main/trader-copycat. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Trader Copycat?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Trader Copycat?

As many as you like. As long as a model supports skills, you can use Trader Copycat with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Trader Copycat AI skill free?

Yes. It is published on GitHub by Jakeschincariol under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇