Early rate$2,400 of senior audit time for $500. Early members keep the rate as it climbs.$2,400 of senior audit time for $500See how
Security research

Security Research.

Public write-ups on the bugs, patterns, and protocols we audit. Engineer-to-engineer, no fluff.

Filter
Showing 12 of 164

All research.

AI Agent Approval Bypass: Audit Checks
Adversarial & AI SecurityMay 13, 2026·12 min

AI Agent Approval Bypass: Audit Checks

Why human approval fails in AI agents when the model still controls risky parameters. Audit checks for coding agents, long-lived agents, and Agentic DeFi.

Read
Pre-audit readiness engine: how Krait catches 12 findings that cost teams 30% more
Zealynx NewsMay 12, 2026·16 min

Pre-audit readiness engine: how Krait catches 12 findings that cost teams 30% more

Krait is the Zealynx pre-audit readiness engine. Catch the 12 finding categories that inflate smart contract audit quotes by 30% — locally, free, before booking.

Read
OWASP ASI04 Explained: Agentic Supply Chain Attacks
Adversarial & AI SecurityMay 12, 2026·12 min

OWASP ASI04 Explained: Agentic Supply Chain Attacks

OWASP ASI04 (Agentic Supply Chain Vulnerabilities) explained: MCP Impersonation, malicious tools, trojanised connectors. Real CVEs, attack patterns, mitigations.

Read
The Web3 Founder's Survival Guide: Everything No Engineering Curriculum Teaches
Industry and ComplianceMay 11, 2026·16 min

The Web3 Founder's Survival Guide: Everything No Engineering Curriculum Teaches

Most Web3 founders are technical and ship great code, then watch the protocol die to non-code problems. Tokenomics, fundraising, governance, regulatory, GTM — this is what kills teams.

Read
MCP Vulnerabilities 2025-2026: 16+ CVEs & Breach Index
Adversarial & AI SecurityMay 8, 2026·14 min

MCP Vulnerabilities 2025-2026: 16+ CVEs & Breach Index

Complete MCP vulnerability index: 16 disclosed breaches and 14+ CVEs since April 2025 across Anthropic, Cursor, Postmark — with OWASP ASI04 patterns. Updated weekly.

Read
How to Build Your Own AI Auditor Agent (Interactive Guide, Multiple Paths)
Adversarial & AI SecurityMay 7, 2026·15 min

How to Build Your Own AI Auditor Agent (Interactive Guide, Multiple Paths)

Build an AI auditor agent that actually works. Multiple paths modeled after the best existing tools, benchmarked against 118 real Code4rena findings.

Read
Supply Chain Attacks in Web3 — From NPM to Protocol Exploits
Adversarial & AI SecurityMay 6, 2026·12 min

Supply Chain Attacks in Web3 — From NPM to Protocol Exploits

Complete guide to Web3 supply chain attacks with 5 vectors, real incidents, and actionable checklist.

Read
JailBreak Technique: Misguided Attention Using Trolley Problem
Adversarial & AI SecurityMay 5, 2026·13 min

JailBreak Technique: Misguided Attention Using Trolley Problem

Learn the Act of Jailbreaking an LLM using Paradox-Inspired Techniques

Read
OWASP Smart Contract Top 10 2026: changes and audit guide
Industry and ComplianceMay 4, 2026·29 min

OWASP Smart Contract Top 10 2026: changes and audit guide

The 2026 OWASP Smart Contract Top 10 elevated business logic to #2 and added proxy & upgradeability at #10. Here's what changed and how to audit.

Read
Flash loan attacks: anatomy of nine-figure DeFi exploits
Web3 Attack VectorsApr 29, 2026·37 min

Flash loan attacks: anatomy of nine-figure DeFi exploits

How flash loans amplify oracle, donation, and reentrancy bugs into $200M+ DeFi exploits — Cetus, Penpie, KyberSwap, UwU Lend case studies plus defenses that hold.

Read
Shadow Audits: How to Learn Web3 Security by Breaking Real Protocol Forks
Zealynx NewsApr 29, 2026·15 min

Shadow Audits: How to Learn Web3 Security by Breaking Real Protocol Forks

Shadow audits replay past public security contests on real protocol forks. Learn Web3 security the way top auditors did: reviewing graded contests, fast.

Read
Social Engineering Models: The Polite Approach
Adversarial & AI SecurityApr 29, 2026·12 min

Social Engineering Models: The Polite Approach

Discover how LLMs Threat detection rate can be reduced for Successful Injection

Read