Sentinel Protocol ICO Analysis

LedgerFund ICO Analysis
LedgerFund
Published in
6 min readApr 6, 2018

ICO Token: UPP
ICO Token price: 1ETH =5000 ETH
Hard cap: $22mn
Raised in private sale: $11mn (30% bonus and lockup 6 months)
Token for Presale: 87,500,000 (15%bonus and lockup 6 months)
Token for crowd sale: 39,000,000
Total supply: 500,000,000

PROJECT OVERVIEW:
Security intelligence platform for blockchain (SIPB), this includes a wide variety of security features, from anti theft, malformed transaction prevention, unknown threat prevention, transaction traceability etc.

The protocol collects hacking reports, which are then delegated to “The Sentinels”, a group of trusted security experts, who confirms the incident and registers the case information to the Threat Reputation Databse (TRDB). The Sentinels are rewarded for participation under the Delegated Proof of Stake (DPOS) model with Sentinel Points (SP), which can later be redeemed for its native currency UPP.

Sentinel Protocol also introduces their own S-wallet, a secure wallet with built in machine learning AI, Fraud Detection System (FDS) and built-in security features such as filtering scam addresses and detecting abnormal behaviour, leveraging TRDB for always up-to-date threat definitions. Sentinel Protocol utilises the strength of decentralised and distributed computing, running extensive tests and simulations in their D-Sandbox (Distributed Malware Sandboxing) environment, effectively cutting cost down compared to the traditional expensive dedicated virtual machine environments. Sentinel Protocol can be integrated seamlessly to any client wallets and exchanges, leveraging its collective intelligence to protect users from malicious threats.

Security Intelligence Platform for Blockchain (SIPB, or Sentinel Protocol) has the following unique security features:
● Threat reputation database (TRDB)
● Machine learning (ML) engine integrated security wallet (S-Wallet)
● Distributed malware analysis sandbox (D-Sandbox)

Threat Reputation Database (TRDB)

Those who have used anti-virus softwares for the internet should be familiar with definition files, each commercial virus scanners maintain their own virus/threat database and end-users need to update on a periodic basis for the latest definition. These databases are not only centralized making them more vulnerable to manipulate but more importantly never complete as vendors lack any incentive to collaborate. Sentinel Protocol aims to create a global alliance for threat information exchange, with built-in incentives for each vendor to contribute to the TRDB.

Machine Learning Engine Integrated Security Wallet (S-Wallet)

Sentinel Protocol’s secure wallet (S-wallet) is similar to an “antivirus enabled wallet”, leveraging collective intelligence harvested by the community and security experts pooled in the TRDB. S-wallets can detect suspicious executions, these can be in the form of blacklisted wallet addresses, scam sites or fradulent activities detected by their Fraud Detection System(FDS), driven by their proprietary machine learning engine. S-wallet also analyzes the threat tendency and history to proactively respond to unknown threats (known as zero-day attacks) that traditional antivirus softwares can’t do before the next software update. This is an AI driven engine, continuously learning from the client side Sentinel Wallet to create model behaviors.

Let’s create a use case to see how this might work in real life. Say you’re about to participate in an ICO, and ICOs these days come with scams by default. There’s a telegram DM popped up on your phone, with a contribution address that will expire in 3 min, it’s been a busy day that you’re not keeping up with the latest update and you are about to FOMO in.

Now when you enter the contribution address into your S-wallet, ready to send the transaction, the send button suddenly turns red with (13 SCAMS REPORTED), you know S-wallet just saved your earned tokens.

Distributed Malware Analysis Sandbox (D-Sandbox)

D-Sandbox is similar to traditional sandbox, a test environment (virtual machines) to run unverified programs. In D-Sandbox environment (decentralised nodes), potential threats are submitted and analysed thoroughly via collective intelligence, at significantly lower costs of computing resources and infinitely scalable.

TOKEN UTILITY:
Thus, Sentinel Protocol has a circulating cryptocurrency named UPP (Uppsala) for the use of the goods and services provided by Security Intelligence Platform for Blockchain (SIPB) and SP (Sentinel Points) for the staking value of The Sentinel Protocol’s reputation.

Sentinel Points is an incentive system where users are rewarded with Sentinel Points by doing Proof Of Protection.

Some form of Proof Of Protection includes: — Various Cybersecurity activities like reporting a scammer’s address or IP, validating report etc — Validation report done by Sentinels — Doing of D sandboxing computation Sentinel Points will then be able to convert to UPP tokens.
The rates are to be revealed in the future. Even though Sentinel Points can be converted to UPP tokens, UPP tokens cannot be converted into Sentinel Points.

UPP tokens is the currency to use the services provided by Sentinel Protocol. Some services are: Leveraging the Machine learning function in S wallet — Running of D sandboxing — In opening a case for detailed cyber forensic service, consultancy, vulnerability assessment, and/or other activities requiring the Sentinel Protocols’ help.

TEAM, ADVISORS AND VC: (Strong)
Team: Strong with work experience in companies like Darktract, Penta Security Systems.

Patrick Kim, Founder & CEO — Cyber Technology Specialist at Darktract (an AI company for cyber security), former System Engineer at Fortinet, Services Consultant at Cisco.

HM Park, Co-Founder & Head of Operations — Major Account Manager at Darktrace, Terriroty Account Manager at F5 Networks, Global Account Manager at Penta Security Systems, and Backend System Programmer at Webcash.

John Kirch, Chief Evangelist — Regional Director at Darktrace, Director at Nominum, Vice President at Penta Security Systems. Began his career at Bank of America as Money Tracking Systems Analyst.

Advisors : (Strong)

Simon Kim: CEO at HASHED
John Ng
:Partner at Signum Capital
JH Kim:
Foundation Council at ICON
Wong Lee Hong :
Executive Advisor at Kyber Network

VC: (Strong VC support)

TOKENOMICS:
Good Tokenomics with total market cap of $35mn.
Token will increase with 20 vesting time details of which will be published in future as reported by the team.Inflationary token supply is not a good thing.

ROADMAP:
Testnet launched on March 2018

Public Beta release in June 2018 and Mainet launch in July, 2018
November 2018
: Machine learning engine beta test
December 2018: Machine learning engine feature release (auto report applied) beta / Distributed sandbox (D-sandbox) release
2019: Machine learning based Fraud Detection System (FDS) release into mainnet

GITHUB:

No Github available

MARKETING: (Weak Marketing)
Twitter followers: 1.5k
Telegram:5k
Reddit:35
Medium: 114
FB: 130
Youtube: https://www.youtube.com/watch?v=gOF0QljFyHQ (play at 3hour 15min for demo.

PROS/CONS:

PROS: No Competition in the project and will have early mover advantage, Strong team and advisor with good VC support. The Team was also the co-winner of the Kyber Network Developer Competition.

CONCERNS:
With no Github and weak Marketing. The project relies on network effect to create a rich database of threats. The protocol’s usefulness is limited if not a lot of people report scam addresses/transactions which is big concern.
Token supply will also increase in future.

CONCLUSION:
We believe the idea is new but lacks blockchain integration and have a centralised approach with sentinels.
We would rate this ICO 5/10.

We, at Ledgerfund run one of the biggest Crypto Hedge Funds in the world and give returns in bitcoins. To know more about how you can invest your bitcoins in this fund, refer to our website: https://ledgerfund.io

--

--