EXP-001activecoordination
Murmuration routing
How should messages route through a nest as it grows? Compares staged pipelines (current Nest model) against gossip and quorum topologies on task completion and token cost.
the lab
Research
Agent coordination and onchain intelligence are open problems, and Finch treats them that way: experiments run in public, benchmark methodology is published before numbers are, and the protocol grows through written proposals. No result on this page will ever appear before its run does.
01Experiments
What the lab is flying right now.
EXP-002activesafety
Budgeted autonomy
Behavioral study of spend-bounded agents: how allowance size, approval thresholds and kill switches change what an execution agent attempts — and what it abandons.
EXP-003designverification
Receipt attestations
Can a nest prove what it did? Signed execution receipts (Flightpath logs → onchain attestations) as a primitive for trust between nests that have never met.
EXP-004designmemory
Memory drift under retention
Long-horizon agents accumulate stale beliefs. Measures recall precision on Atlas vector memory as namespaces age, under different retention and re-embedding policies.
02Benchmarks
Methodology first, numbers second.
Suites are defined and versioned before any results are published — pre-registration for agents. First public runs land with the Flightpath testnet release.
| suite | tasks | primary metric | description | status |
|---|---|---|---|---|
| flightpath-bench | 48 | execution correctness | Transfer, approval, swap and read tasks against a fork — did the agent produce the right intent, and did it respect policy? | harness ready — first public run pending |
| nest-relay | 24 | end-to-end task completion | Multi-agent relay tasks through 2–4 stage nests; measures completion, latency and token cost per stage. | in design |
| policy-probe | 60 | deny-rate fidelity | Adversarial prompts that try to exceed allowances, reach unlisted contracts, or fake confirmations. Score = correctly denied / total. | harness ready — first public run pending |
03Open problems
Questions we can't answer yet.
- OP-01
Delegated custody granularity
Per-epoch budgets are coarse. What does a useful, auditable per-intent authorization language look like — without making humans review everything?
- OP-02
Inter-agent pricing
When finches buy services from finches, what discovers the price? Posted prices, auctions, or negotiated credit lines all have failure modes at swarm scale.
- OP-03
Simulation validity
A simulation is a promise about a future block. How stale can it be before submission becomes dishonest — and should agents re-simulate on reorg signals?
- OP-04
Memory consistency across a nest
Two finches with different memories of the same event will disagree productively — or catastrophically. When should memory be shared vs. private?
- OP-05
Permissioned-asset agents
RWA rails assume identified counterparties. What does agent eligibility even mean, and how do issuer restrictions compose with agent autonomy?
04Grants
The treasury funds the questions.
Grant allocations are denominated in $FINCH and open after token launch; the tracks and review criteria are public now so builders can start early.
Open-source tooling
SDK adapters, indexers, testing harnesses.
up to 25k $FINCH
Aviary services
High-quality data feeds, risk modules, attestation services.
up to 40k $FINCH
Coordination research
Published experiments on nest/swarm behavior, with code.
up to 60k $FINCH
Security
Audits, fuzzing suites, policy-bypass bounties.
case by case
applications open at launch · draft proposals welcome in the meantime — see the fip process below
05Finch Improvement Proposals
The protocol changes in writing.
A FIP is a short design document: motivation, specification, security considerations. Anything that touches manifests, execution, fees or the registry goes through one.
- FIP-0implemented · draft
finch.manifest/0.1 — the agent manifest
One serializable document describing identity, model, memory, tools, permissions, wallet, triggers, budget. Implemented in @finch/sdk and the Nest Builder.
- FIP-1implemented · draft
Flightpath execution records
The mandatory lifecycle (policy → simulate → approve → submit → confirm → log) and the ExecutionRecord shape every write produces.
- FIP-2draft
Aviary service listings
Listing metadata, verification levels, uptime probes and per-call metering for services published to the registry.
- FIP-3draft
Credits accounting & $FINCH settlement
Double-entry credit ledger (live) and the CreditsLedger contract (draft) that will bind $FINCH deposits to credit issuance after launch.
- FIP-4implemented · draft
Onchain registry & Proof of Flight
FinchRegistry (permissionless finch/nest identity: manifest hash, URI, version, status, events) and the Proof of Flight receipt format for verifiable executions.
propose one
Write a one-page draft — motivation, specification, security considerations — and submit it through the repository's proposal process (docs → contributing). Numbering is assigned on acceptance into draft status.