What is Prometheus Protocol?

Prometheus is a decentralized, AI-powered cybersecurity protocol built on Kaspa. It is building a data-minimal global threat detection swarm with no foundation or pre-mine. The contract design has no repository-controlled emergency-stop entrypoint; rollout verification remains in progress.

What is the PROM token?

PROM is the earned-only reward and governance token with 0% pre-mine. It is minted only through consensus-accepted security contribution paths. Canonical Guardian reputation is separate Kaspa L1 state, not a PROM balance.

How does Prometheus detect threats?

Target lifecycle: Phi-3-mini detects locally and emits an approved, data-minimal proof-bound report. LLaMA 3 analyzes validated observables and generates a YARA rule. Validators vote via Commit-Reveal, and accepted rule state is anchored on Kaspa L1. The current verified ThreatHint v1 path is deliberately non-actionable.

When does Prometheus launch?

Kaspa Toccata runtime and release gates pass. The non-promotable H-001 testnet-10 canary has confirmed public funding and a byte-identical schema-v2 signing request rebuilt from exact main 205e1ca with live UTXO revalidation; external signing, verified broadcast, confirmation, and independent evidence still gate the canary. Full rollout additionally requires the remaining contracts, metrics-oracle transaction, and exact-commit release hardening.

Kaspa Toccata · Deploy verification active

The security the world
never controlled.

Prometheus is a decentralized AI threat intelligence protocol built on Kaspa. No foundation or pre-mine. Contract design has no repository-controlled emergency-stop entrypoint; rollout verification remains in progress.

Read the Whitepaper FAQ & Docs GitHub

Pre-mine · Founder allocation · Investor tranche

<60s

Threat detection to immutable on-chain rule

100%

Open source · MIT License · Community owned

01 / Protocol

Target: on-chain in under 60 seconds.

The target path from anomaly to validated Kaspa rule after proof, observable, consensus, and rollout gates pass. Current verified ThreatHint v1 is deliberately non-actionable.

t=0s

T = 0 s

Light Client

Target: Phi-3-mini detects locally; artifact hash and privacy-filtered observable commitment are bound by an approved, precisely scoped proof.

ONNX Runtime
Groth16 ZK-proof
libp2p

t=2s

T = 2 s

Guardian Node

LLaMA 3 aggregates reports. Analyzes entropy, API patterns, CVE correlations. Generates YARA rule with confidence score ≥ 0.85.

LLaMA 3 70B / 8B
vLLM · YARA
Min confidence 0.85

t=15s

T = 15 s

Validator Network

Commit-Reveal voting. Commitment = sha256(vote ‖ salt ‖ block). 10% bond locked. 67% consensus required.

Silverscript
Commit-Reveal
Bond system

t=35s

T = 35 s

Kaspa Layer 1

Rule stored as unique KRC-20 asset. Supply: 1. Immutable. DAGKnight sub-second finality. IPFS CIDv1 reference.

DAGKnight
KRC-20 (supply=1)
IPFS CIDv1

t=36s

T = 36 s

All Clients

P2P broadcast or Kaspa polling. New rule automatically loaded. Local scanner active immediately. Every device protected.

rusty-kaspa
libp2p broadcast
Local YARA engine

02 / Architecture

Three layers.
No center.

Every component is replaceable. Every rule is verifiable. Every decision is on-chain.

1 Kaspa Layer 1

Immutable truth

DAGKnight consensus · 100 BPS
Silverscript smart contracts
KRC-20 rule storage (supply = 1)
Native Groth16 ZK-verification
Sub-second finality
ValidatorStaking · RuleStorage

2 Kasplex zkEVM

Intelligence layer

Complex governance logic
ZK-rollups anchored to L1
Reputation computation
Reward distribution
Dev grant DAO voting
Auto-tuning parameters

0 Off-Chain · Device

Action layer

Phi-3-mini local inference
LLaMA 3 Guardian analysis
Fed-DART federated learning
Local files and paths stay local; bounded metadata is explicit
P2P coordination via libp2p
Deterministic bounded automation; rollout gates remain explicit

03 / Node types

Four roles.
Every device contributes.

01 · Light Client

Light Client

Target: local Phi-3 inference with explicit, bounded reporting. Current v1 sends claim metadata but no concrete IOC and makes no anonymity guarantee.

4 GB RAM · No GPU required
Windows · macOS · Linux · ARM

02 · Guardian

Guardian Node

Runs LLaMA 3. Aggregates threat hints, performs deep analysis, generates YARA rules, earns PROM rewards, and builds separate L1 reputation through accepted proposals.

8B: RTX 4070 Ti+, 16 GB VRAM
70B: 4× A100 / H100

03 · Validator

Validator Node

Stakes KAS. Votes on rule proposals via Commit-Reveal. Provides economic security. Slashed for misbehavior. 7-day exit cooldown.

Minimum 10,000 KAS stake
Standard server · No GPU needed

04 · Honeypot

Honeypot Node

Emulates vulnerable services. Every attack captured and analyzed. First line of zero-day detection. Earns 5% of total PROM emission.

Any internet-exposed server
Higher reward per zero-day find

04 / Economics

Two tokens.
No shortcuts.

KAS provides economic security. PROM is never purchased — only earned through contribution.

KAS

Kaspa native token · Validator staking

Validators stake KAS as collateral. Misbehavior costs real capital. The alignment of incentives is economic, not ideological.

Min stake10,000 KAS · auto-adjusted weekly

Slashing5% simple · 10% double-vote · 20% collusion

Cooldown7 days on exit · no flash exits

SourceKaspa network · 0% pre-mine · fair launch

Liquidity$1B+ market cap · no bootstrap needed

PROM

Prometheus token · Reputation & governance

PROM cannot be bought at launch. It is minted only when a threat rule is accepted by consensus. Every token represents work done.

Year 120,000,000 PROM · −10% per year

Validators40% · for validated rules

Guardians30% · for accepted proposals

Reporters20% · Light Clients + Honeypots

Dev Pool5% · DAO vote only · no foundation

06 / Roadmap

Built in public.
Audited in public.

Sprints, findings, evidence, and remaining gates are tracked in the public repository.

Mar 2026

Sprint 0 — Setup & Testnet

Kaspa Testnet-10 node · repository structure · CI/CD pipeline

Accepted

Mar 2026

Sprint 1 — Silverscript Contracts

6 contracts · 54 tests · ValidatorStaking · GuardianReputation · RuleStorage

Accepted

Mar 2026

Sprint 2–3 — Rust Client & AI

Kaspa RPC · YARA scanner · Phi-3 wrapper · ZK-proof stub · Fed-DART

Accepted

Mar 2026

Sprint 4–5 — Guardian & Voting

vLLM Docker · YARA generator · Commit-Reveal · slashing engine

Accepted

Mar 2026

Sprint 6–7 — E2E & Documentation

Development-stub lifecycle fixture <60s · security tests · audit dashboard · whitepaper; not production evidence

Accepted foundation

Apr 2026 onward

Community workstream

Gitcoin Grants · public repository · external contributors; separate from accepted Sprint 8 Public Site delivery

In progress

Jul 2026

H-001 testnet-10 canary — 96% prepared

runtime gates · public funding · exact-main 205e1ca byte-identical schema-v2 handoff and live UTXO revalidation complete · external signature, verified broadcast, receipt, and evidence remain

Next gate

Readiness-gated

Complete roadmap vision — 44–49%

core rollout 78–82% · reviewed observable extractors/privacy gates · separate v2 wire and statement/relation · approved proof artifacts · privacy-preserving channel · production AI/P2P · clients · operated network · vProgs

In progress

The fire belongs

to humanity,
not to corporations.

No foundation controls the protocol and no repository-controlled emergency-stop entrypoint is planned. Availability still depends on Kaspa, nodes, clients, and network access; no uninterrupted-access guarantee is claimed.

Join on GitHub Read the Whitepaper

MIT License · 0% Pre-mine · 100% Open Source · Community owned

The security the world never controlled.

Target: on-chain in under 60 seconds.

Three layers.No center.

Four roles.Every device contributes.

Two tokens.No shortcuts.

Built in public.Audited in public.

to humanity,not to corporations.

The security the world
never controlled.

Three layers.
No center.

Four roles.
Every device contributes.

Two tokens.
No shortcuts.

Built in public.
Audited in public.

to humanity,
not to corporations.