solidity-auditor

Audits Solidity smart contracts for vulnerabilities using parallel scanning agents and attack-vector references.

1|1|Updated May 21, 2026
One-click install
npx skills add https://github.com/naruto11eth/cryptoskills --skill solidity-auditor-naruto11eth
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: solidity-auditor
Source: https://github.com/naruto11eth/cryptoskills/tree/main/skills/solidity-auditor
Command: npx skills add https://github.com/naruto11eth/cryptoskills --skill solidity-auditor-naruto11eth

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and assets (resource) components.

What problem does it solve? Smart contract bugs can lead to lost funds, but manual security reviews are slow and expensive. This Skill runs an automated security audit of Solidity code during development, surfacing high-confidence vulnerabilities in minutes before every commit. ## Core Features & Use Cases - Parallelized Multi-Agent Audit: Orchestrates four vector-scanning agents against 170 documented attack vectors, then merges and deduplicates findings by root cause. - Deep Adversarial Mode: An optional DEEP mode spawns an adversarial reasoning agent that hunts for logic errors, economic exploits, and cross-function attack paths beyond pattern matching. - Confidence-Scored Reports: Every finding passes a false-positive gate and receives a confidence score, with formatted reports including descriptions, diffs of suggested fixes, and a findings table. - Use Case: Before committing changes to a vault contract, run the audit on src/Vault.sol to catch missing access controls, reentrancy, or rounding errors and get a fix-ready report. ## Quick Start Ask the agent to run a security audit on your Solidity repository, or pass a specific filename like src/Vault.sol to review just that contract.

Frequently Asked Questions about solidity-auditor

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a security audit on Solidity contracts?

Invoke the skill with no arguments to scan all in-scope .sol files in the repo, or pass specific filenames to review only those contracts. Test, mock, interface, and library directories are excluded automatically.

What kinds of vulnerabilities does a Solidity audit detect?

It checks 170 documented attack vectors including reentrancy, access control gaps, oracle manipulation, ERC4626 inflation attacks, signature replay, and proxy storage collisions. A DEEP mode adds adversarial reasoning for logic and economic exploits.

What is the difference between default and deep audit mode?

Default mode runs four parallel vector-scanning agents across the codebase. Deep mode adds a fifth adversarial reasoning agent that hunts for multi-transaction, cross-protocol, and game-theoretic attacks, at higher cost and latency.

What are the limitations of AI smart contract auditing?

Accuracy drops on codebases beyond roughly 2,500 to 5,000 lines, so large projects should be audited per module. AI also struggles with invariant bugs, cross-protocol composability, and off-chain assumptions, so it does not replace a manual audit.

How are audit findings scored and reported?

Each finding passes a three-check false-positive gate, then receives a confidence score starting at 100 with deductions for privileged callers or partial attack paths. Findings are sorted by confidence, and those below the threshold get descriptions without fix sections.