Liquidity has risk

Providing liquidity is not free money. A position can lose value, stop earning fees, change asset composition, or become difficult to exit. Risk varies by market, range, asset, issuer, reference source, and external settlement process.

Those eight are listed separately because they have separate causes, but an LP is exposed to all of them at once and continuously. No configuration eliminates any one of them, and several originate outside the protocol entirely. The point of enumerating them is not to suggest they can be managed away one at a time, but to make it possible to say which ones a specific position is most exposed to.

Impermanent loss is the item most often misunderstood, largely because the name suggests it reverses. It describes the gap between what a position is worth and what the same assets would have been worth if left alone, and it becomes permanent the moment the position is closed. In a binned market it takes a particularly clear form: a position the market has passed through has sold the asset that appreciated, or bought the one that fell, at every level along the way. The fees earned are the compensation for that, and whether they cover it depends on how much trading occurred rather than on where the price ended up.

Risks compound

The risks above are listed separately but rarely arrive that way. One sequence of events can trigger several at once, and the combination is usually worse than any of them read alone.

Nothing in that sequence is a malfunction. Every step is the system behaving as designed, which is why gap risk in tokenized real-world assets is a structural property rather than an edge case.

The scenario is drawn from tokenized equities because the weekend gap makes it concrete, but the pattern is general. Any asset whose price is anchored to something that updates less often than the onchain market trades has this shape, and the size of the exposure is set by how far the anchor can travel while nobody is able to observe it. A tokenized Treasury has a small version of this problem. A private-credit token with a monthly NAV has an extreme one.

Reducing exposure

None of these remove risk. They change how much of it a position carries, and they are decisions made before capital is committed rather than after.

• Know the issuerUnderstand who mints the token, what it entitles the holder to, and how redemption works.
• Size the range deliberatelyA range is a statement about where you are willing to hold each asset.
• Prefer referenced marketsAn asset with reliable external data can be checked against something.
• Watch composition, not entry priceWhat a position holds now determines what withdrawing returns.
• Assume gapsMarkets with closed underlyings can reopen at a materially different price.

None of these five is a hedge, and it is worth being clear about why. A hedge is a second position that offsets the first, and nothing described here does that. These are choices about which risks to accept and in what size, made before any capital is committed. An LP who wants their directional exposure neutralized needs an instrument that does it, and providing liquidity is not one.

Risk is per market, not per protocol

Two markets running the same contracts can carry very different risk. A tokenized equity with a live reference and deep liquidity is not comparable to a private-credit token with a monthly NAV and few buyers. Any statement about safety should name the market it applies to.

That is also the right way to read any claim about Silo's safety, including claims made in this documentation. Statements about the contracts apply protocol-wide. Statements about liquidity, reference quality, redemption, or issuer standing apply to one market and cannot be carried across to the next, even where the two run identical code and sit side by side in the same interface.

What Silo does not control

Several of the risks above originate outside the protocol entirely. Silo can surface them, restrict activity around them, and describe them honestly, but it cannot remove them.

The line between what the protocol controls and what it merely observes matters when something goes wrong, because the two have different remedies and neither is the protocol's to apply. If a reference feed fails, Silo can restrict execution; it cannot produce a correct price. If an issuer suspends redemption, Silo can keep operating a market in the token; it cannot make the token redeemable. An LP's recourse in those cases lies with the party that failed, which is why knowing who that party is forms part of the decision to deposit at all.

Questions worth answering before depositing

These are the questions whose answers determine most of a position's outcome. An LP who cannot answer them for a specific market does not yet know what the position is exposed to.

• Who issues this token, and what does holding it entitle you to?
• Can it be redeemed, by whom, and over what period?
• What is the reference source, how often does it update, and what happens when it fails?
• When is the underlying market open, and how far has this asset gapped on reopening before?
• How much depth sits near the current price, and how much of it would be yours?
• Which asset will the position hold if price leaves the range, and is holding it acceptable?
• What would make you close the position, and would you be watching when it happened?

The last question is the one most often skipped and the most predictive. A position is only as well managed as the attention actually available to it, and attention is not evenly spread across a week. A range requiring daily adjustment, held by someone who checks it weekly, is in practice a wider and far less deliberate position than the one that was intended.