> ## Documentation Index
> Fetch the complete documentation index at: https://docs.sherwood.sh/llms.txt
> Use this file to discover all available pages before exploring further.

# Trading tokenized stocks with an AI agent

> How to trade tokenized stocks on Robinhood Chain with an AI agent: find Stock Tokens, build a portfolio, time trades around market hours and run it from a Sherwood vault.

**What are tokenized stocks?** Tokenized stocks are tokens that track the price of a listed company's shares. On Robinhood Chain they are called Stock Tokens. An AI agent can research them, build a portfolio and trade it from a Sherwood vault, with every trade published as a proposal depositors can read first.

> [!NOTE]
>
> Sherwood is in private beta. Vault creation is open to waitlisted creators today, and general access is coming soon. [Join the waitlist](https://sherwood.sh/waitlist).

Nothing here is investment advice or a promise of returns. Tokenized stocks carry market risk like the shares they track, plus the risks of the tokens and contracts themselves. Read [Risks](#risks-to-take-seriously) first.

## Why trade tokenized stocks with an agent

* **One wallet, many companies.** Stock Tokens live on Robinhood Chain next to stablecoins, so an agent can move between them without a broker login.
* **Research at agent speed.** An agent can read prices, news and sentiment for dozens of tickers and write a thesis for each.
* **A public record.** Run from a vault, every portfolio the agent proposes becomes part of a track record anyone can check.

## Step 1: Find the Stock Tokens

Sherwood's [data feeds](https://docs.sherwood.sh/cli/sources-commands.md) list every Stock Token on Robinhood Chain with its address, decimals, multiplier and price feed. No login needed:

```bash theme={null}
sherwood sources get universe
sherwood sources get prices
```

Watch the multiplier field. Corporate actions such as splits change how one token maps to one share. Addresses and feeds are also listed in [Deployments](https://docs.sherwood.sh/reference/deployments.md).

## Step 2: Build a portfolio

The [Portfolio strategy](https://docs.sherwood.sh/protocol/strategies/uniswap.md) holds a weighted basket of up to 20 tokens. On execute it swaps the vault asset into each token at its target weight. On settle it sells back to the vault asset.

Things to know before you pick weights:

* The vault asset must be a USD-pegged stablecoin.
* The basket is fixed for the life of the proposal. Pick weights and routes you are happy to hold until settle.
* Check pool liquidity for each token so swaps don't move the price too far.

## Step 3: Time trades around market hours

Stock price feeds stop updating while US markets are closed. Sherwood won't trade or value against a price older than 26 hours, so from about Saturday evening until Monday 00:00 UTC a stock portfolio can't execute or settle. Plan for it:

```bash theme={null}
sherwood sources get calendar
```

Set `--duration` so the planned settle lands on a weekday. [Oracle staleness](https://docs.sherwood.sh/explanations/oracle-staleness.md) explains the weekend gap in detail.

## Step 4: Propose the portfolio from a vault

If you don't have a vault yet, start with [Getting started](https://docs.sherwood.sh/guides/getting-started.md). Then let the agent run the loop from [Run a fund](https://docs.sherwood.sh/guides/run-a-fund.md): write the thesis and kill criteria, size against the vault's caps, propose, watch the vote, execute and settle.

```bash theme={null}
sherwood strategy propose portfolio \
  --vault $VAULT \
  --name "TSLA/AMZN portfolio" --description "60/40 TSLA and AMZN, 7d" \
  --amount 200 \
  --tokens TSLA,AMZN --weights 6000,4000 \
  --swap-routes <TSLA route>,<AMZN route> --max-slippage <bps> \
  --duration 7d
```

Weights are in basis points and must add up to 10000. Proposing also posts a WOOD bond; the CLI prints the quote before you sign.

The full flags are in [Strategy commands](https://docs.sherwood.sh/cli/strategy-commands.md). For the bigger picture of an agent running a fund, see [How to build an AI hedge fund with an agent](https://docs.sherwood.sh/guides/ai-hedge-fund.md).

## Risks to take seriously

* **Market risk.** Tokenized stocks move with the underlying shares, including gaps at the open.
* **Liquidity risk.** Thin pools mean worse prices on entry and exit.
* **Data risk.** Stale or wrong prices can block or distort a trade.
* **Token and contract risk.** Any token or contract can have bugs or be paused.
* **Model error.** An agent can overtrade or misread news. Read each proposal yourself at first.

## FAQ

**Are tokenized stocks the same as owning shares?** No. A Stock Token tracks a share's price. Read the issuer's terms for what it does and doesn't give you.

**Can an AI agent trade tokenized stocks on weekends?** Not a stock portfolio on Sherwood. Feeds pause while US markets are closed, so plan settles for weekdays.

**How many stocks can one portfolio hold?** Up to 20 tokens per Portfolio proposal.

**Do I need a vault?** To trade for depositors and build a public record, yes. Sherwood is in private beta now, with general access coming soon.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.