Skip to content

Milestones

List view

  • M4 · KERI identity

    Actors become identities: AID-bearing ActorIds, the Kevery-equivalent runtime (escrow storage, retry timers, cues) wired to cesr sans-io verdicts, and key-custody backends. cesr owns ALL pure KERI logic; bombay owns the IO/runtime half. Depends on M1 (Zenoh) + M2 (nexus adapter) + cesr KERI-core milestone.

    No due date
    •0/1 issues closed
    0% complete1 open0 closed

  • M0 · Pre-flight (de-risk before the fork)

    Cheap de-risking BEFORE forking 19k LOC: prove the Zenoh assumptions in a throwaway spike, decide fork strategy, pin the Zenoh feature matrix. Everything in M1+ rests on these.

    No due date
    •3/3 issues closed
    100% complete0 open3 closed

  • M7 · De-handroll (crate adoption)

    Replace kameo hand-rolled code with best-in-class Rust crates — but only the LOCAL core that survives the Zenoh migration. Zenoh already deletes the libp2p remote layer (DHT registry, messaging, behaviour, swarm) and its error/From boilerplate, so those are not carded here.

    No due date
    •6/6 issues closed
    100% complete0 open6 closed

  • M3 · Novel Zenoh-native features

    Capabilities Zenoh gives that kameo lacks: wildcard/hierarchical key-expr group addressing, queryable + persistent actor state via storages, liveliness-based supervision and membership.

    No due date
    •0/28 issues closed
    0% complete28 open0 closed

  • M1 · Foundation: Zenoh remote layer

    Fork kameo and replace the libp2p remote module with a thin layer over a Zenoh Session. Generic, domain-agnostic core. ~4,200 LOC of plumbing collapses to ~800.

    No due date
    •18/38 issues closed
    47% complete20 open18 closed