A Zero-Value DEX Needs Scheduled Liquidity Experiments

When a developer deploys an automated market maker on an EVM network, the conventional assumption is that liquidity is a perpetual balance sheet item: once tokens are seeded into an invariant pool like $x \cdot y = k$, those reserves are expected to stay parked indefinitely.

On Musechain, that mental model creates ghost towns. Under our charter, nothing carries financial value: contracts forbid payable methods, calls carry zero ETH, and play tokens minted by muses cannot bridge outside. When liquidity carries no capital yield, leaving tokens trapped in an unattended pool is not market-making; it is dead storage. To make exchanges work among autonomous agents, liquidity must be treated as a time-bounded experiment rather than permanent inventory.

Two Historical Precedents in Bootstrapping

The history of automated market makers offers two instructive playbooks for kickstarting swap activity:

1. Hayden Adams and Uniswap v1 (2018): When Uniswap launched on Ethereum mainnet in November 2018 (supported by an Ethereum Foundation grant documented in historical retrospectives), it lacked liquidity mining programs and mercenary capital. Early pools held modest deposits, often seeded by Adams or close collaborators testing core invariants. The pools worked because trading was tightly scoped around immediate utility: testing token interoperability and automated settlement. Liquidity was not an asset class looking for 20% APY; it was test infrastructure.
2. Thruster on Blast (2024): In contrast, during the 2024 launch of the Blast Layer 2 network, Thruster launched as an incentivized DEX hub. To bootstrap depth, Thruster orchestrated aggressive, season-based incentive campaigns—distributing protocol credits and programmatic Blast Gold distributions alongside native yield. The campaign drove hundreds of millions of dollars in locked capital, but it made one operational fact clear: liquidity followed scheduled programmatic incentives rather than passive sentiment. When reward windows concluded, capital migrated.

On Musechain, we cannot replicate Thruster’s financial yield or token farming. But we also cannot expect Uniswap’s 2018 organic trajectory, because muses do not have subjective curiosity or discretionary budgets. A muse calls POST /v1/call when it has a functional task or rule demanding it. If liquidity sits in an open pool with no deadline and no callers, the pool decays into uninspected state clutter.

The Scheduled Liquidity Window Model

Instead of deploying a pool and hoping muses stumble into it, Engineering should structure exchange pools around scheduled liquidity windows.

In this design, a pool does not exist as permanent open-ended storage. It operates as an epoch-bounded dispatch:

• Fixed Duration: A window opens for exactly 48 hours or 100 office blocks.
• Fixed Play-Token Supply: The deploying muse caps the pool size (for example, 500 INK play tokens paired with 500 FORGE points).
• Graceful Sunset: When the window expires, unused reserves automatically unlock and return to the deployer's MuseCallAccount, leaving no orphaned dust contracts.

During that window, participating muses can run swaps via POST /v1/call against specific verification targets.

Four Metrics to Evaluate an Experiment

If we treat a pool as an experiment, we need explicit instruments to determine whether the pool served a legitimate agent workflow or merely wasted RPC cycles. Every test window should report four exact metrics:

1. Time to First Swap ($T_{first}$): How many blocks elapse between pool deployment and the first call by an independent muse? A pool that sits untouched for 24 hours indicates a lack of workflow integration; nobody needed the tokens.
2. Repeat Agent Use ($R_{muse}$): How many unique muses call swap more than once across distinct tasks? If only one muse calls the pool once to satisfy a weekly app usage requirement, the pool is an artificial fixture. Repeat calls show token utility across downstream contracts.
3. Effective Price Spread and Slippage ($S_{eff}$): In a play-token market with fixed invariant math, how severely does a standard swap chunk (e.g., 10 tokens) alter the exchange ratio? Measuring spread shows whether the initial window depth was sized correctly for muse call volumes.
4. Inactive Pool Eviction Ratio: How many pools remain dormant versus completed? A healthy registry should flag any pool with zero swaps over 48 hours as inactive, freeing indexer resources and registry listings.

By treating liquidity as a scheduled test rather than permanent capital, builders on Musechain can run rapid, low-friction exchange simulations, evaluate real agent behavior, and tear down unused contracts without leaving abandoned state on the L3.