Reading the trails
- --foragers
- --under colony
- --30d realized
- --trails active
- 0.31anchor
- not yet deployedprogram
1 of 6 reported. Realized figures only.
1 of 6 reported. Realized figures only.
Reading the trails
brood vault
The Brood Vault is the single entry point for capital and the accounting root for every share. From there the colony routes it outward across dozens of isolated sub-accounts, and the routing is decided by realized performance rather than by a manager.
This page is the vault side of that mechanism: where a deposit goes, how the weights that move it are computed, what the rebalance costs, and where the protection stops.
the deposit path
Base asset enters one vault and mints shares against net asset value. There is no per-agent deposit and no per-agent withdrawal, so a depositor never has to pick which forager to back.
Ten percent of value under colony is held out as the exploration reserve and is never allocated by pheromone. The remaining ninety percent is the main pool that the weights act on.
At each epoch boundary the vault moves net deltas into per-forager sub-accounts. Capital can leave a sub-account only to a whitelisted venue or back to the vault, never to an arbitrary address.
the weighting pipeline
The same four steps run every epoch, in this order, on the committed on-chain inputs. Anyone holding those inputs can recompute the result and get the same weights.
Each active forager's pheromone is divided by the sum across the roster, giving a raw share.
Shares have to sum to one before anything can be bounded, so the raw normalization comes first.
Any share above twenty percent is cut back and the excess spread proportionally over the rest, repeating until nothing exceeds the cap.
Concentration is the systemic risk on the allocation side, so it is bounded before capital moves rather than after.
Any share under three percent goes to zero. That forager is demoted to Scout and its capital is withdrawn. It is a demotion threshold, not a floor.
A decayed trail is demoted instead of propped up at a floor, because a minimum weight would fight the whole mechanism.
Surviving weights are rescaled to sum to one and become the epoch's targets.
Capping and dropping both leave the vector short of one, so the survivors are rescaled into the epoch targets.
the edge case
N_active * w_max < 1effective_cap = max(w_max, 1 / N_active + margin)Weights cannot sum to one while every weight stays at or below the cap unless there are enough active foragers to carry them. When there are not, the cap relaxes to the value above, and any capital that still cannot be placed under it stays in the vault and is reported as un-deployed reserve.
It is never hidden, and it is never silently over-concentrated into whichever forager happens to be on top. A shortfall of foragers is a number on the vault, not a quiet change to the risk you took.
There is a second consequence. At exactly N_active = 1 / w_max the cap forces every weight to the cap and pheromone can no longer express any preference at all. With the published cap of 20% that floor is five active foragers, so the colony wants comfortably more than five before weighting means anything.
epoch and rebalance
Rebalancing costs slippage and gas, and that cost is socialized across the vault. The band and the cap keep it small; they do not make it zero.
the limit of isolation
Isolation caps blast radius. It does not make a pooled fund immune to its own aggregate result: net asset value is the sum of all sub-accounts, so a large loss in one forager still lowers every depositor's share by that forager's slice.
Sub-account isolation is enforced by the program, not promised by convention, and it does exactly one job: it stops one forager from reaching another one or from touching the un-deployed principal. It does not move a loss off the pool. The concentration cap bounds how large any single slice can be; the bond and the cache absorb part of what is lost; the remainder lands on share value.
Agents lose money. Allocation is not a forecast.