Skill 08 security audit

Skill vaquarkhan/web3-agent-skills/skills/skill-08-security-audit

Production-grade Agent Skills for Web3 AI agents: 15 workflows, 6 MCP servers, DeFi/NFT/compliance guardrails, and VS Code/JetBrains installers.

Install

npx -y skills add vaquarkhan/web3-agent-skills --skill skill-08-security-audit

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Performs Web3 security analysis including vulnerability detection, rug pull checks, and exploit monitoring. Use before interacting with unknown contracts or when auditing smart contract code.

SKILL.md

2.5 KB, as published. Nobody here has run it

Security Audit

When to Use

Evaluating unknown smart contracts before interaction
Detecting rug pull and honeypot patterns
Monitoring for active exploits and protocol incidents
Reviewing contract code for common vulnerabilities
Pre-deployment security checklist

Prerequisites

MCP: blockchain-rpc-server, defi-protocol-server
Knowledge base: knowledge-base/security/

Workflow

1. Automated Rug Pull Checks

Run before any token swap or liquidity add:

Check	Red Flag
Ownership	`owner()` can mint unlimited tokens
Liquidity	LP not locked or unlock imminent
Honeypot	Sell simulation reverts
Hidden fees	Transfer tax > 10% or modifiable
Proxy	Unverified implementation, no timelock
Blacklist	`isBlacklisted` or pause on transfers

Use eth_call simulation: attempt buy → approve → sell in single trace.

2. Vulnerability Patterns

Scan for OWASP Smart Contract Top 10:

Reentrancy — external calls before state updates
Access control — missing onlyOwner on critical functions
Oracle manipulation — spot price without TWAP
Flash loan attacks — single-block price dependency
Integer issues — unchecked math (pre-0.8.0)
Delegatecall — untrusted implementation in proxy
Front-running — no slippage/deadline protection

See knowledge-base/security/common-vulnerabilities/ for patterns.

3. Exploit Monitoring

Monitor Rekt News, DeFiLlama hacks dashboard
Check contract against known exploit DB
Verify protocol TVL hasn't dropped > 20% in 24h

4. Code Review Checklist

Uses Solidity ≥ 0.8.0 with overflow checks
External calls use checks-effects-interactions
No selfdestruct or delegatecall to user input
Timelock on admin functions (> 24h for DeFi)
Events emitted for state changes
Formal verification or audit report available

Verdict Format

RISK: LOW | MEDIUM | HIGH | CRITICAL
FINDINGS: [list with severity]
RECOMMENDATION: PROCEED | CAUTION | BLOCK

Block interaction when RISK = CRITICAL or honeypot detected.

References

knowledge-base/security/audit-patterns/
knowledge-base/security/exploit-db/