Install
1
Install the plugin$
npx claudepluginhub ivantorresedge/molcajete.aiWant just this command?
Then install: npx claudepluginhub u/[userId]/[slug]
Description
Perform comprehensive security audit using multiple tools
Command Content
IMPORTANT: Immediately use the Task tool with subagent_type="sol:auditor" to delegate ALL work to the auditor agent. Do NOT do any analysis or work in the main context.
Use this exact prompt for the agent: "Execute the security audit workflow following these steps:
-
Identify Contracts in Scope
- Use Glob to find all contracts:
src/**/*.solorcontracts/**/*.sol - List all contracts that will be audited
- Understand contract purposes from NatSpec and code structure
- Use Glob to find all contracts:
-
Run Automated Security Analysis
- Follow security-audit skill methodology
- Run Slither for static analysis:
slither . --json slither-report.json slither . --detect reentrancy-eth,unchecked-transfer,dangerous-strict-equalities - Run Mythril if available:
myth analyze contracts/*.sol -o json > mythril-report.json - Run Echidna for fuzzing if configured:
echidna-test contracts/MyContract.sol --config echidna.yaml - Run Foundry fuzz tests:
forge test --fuzz-runs 10000
-
Perform Manual Code Review
- Use security-audit skill checklists:
- Common vulnerabilities checklist (reentrancy, access control, integer issues, external calls, oracle manipulation, front-running, DoS, timestamp dependence, delegatecall, signature replay)
- Access control checklist (admin protection, role checks, privilege escalation, multi-sig, timelock)
- Token-specific checklist if applicable (ERC compliance, approval race, transfer checks, fee-on-transfer, decimals)
- DeFi-specific checklist if applicable (oracle manipulation, flash loans, price staleness, slippage, liquidation)
- Upgrade-specific checklist if upgradeable (storage layout, initializers, authorization, storage gaps, upgrade testing)
- Review code manually for logic errors and business logic issues
- Use security-audit skill checklists:
-
Classify All Findings by Severity
- Critical: Direct loss of funds, contract takeover, unauthorized state manipulation
- High: Significant loss under specific conditions, privilege escalation, critical function DoS
- Medium: Partial loss of funds, temporary DoS, front-running opportunities
- Low: Gas optimizations, code quality issues, minor edge cases
- Informational: Best practice recommendations, documentation improvements, code clarity
-
Generate Comprehensive Audit Report
- Follow security-audit skill report template
- Include:
- Executive Summary (overall assessment, critical findings, recommendations)
- Scope (contracts audited, tools used, audit date)
- Findings (grouped by severity with file, lines, description, proof of concept, recommendation for each)
- Summary (counts by severity)
- Recommendations (prioritized actions)
- For each finding provide:
- Severity classification
- File and line numbers
- Detailed description
- Proof of concept code if applicable
- Specific, actionable remediation recommendation
-
Create Summary Report
- Display summary with counts:
Security Audit Summary: - Critical: X - High: X - Medium: X - Low: X - Informational: X - Highlight critical and high severity findings
- Provide next steps and recommendations
- Display summary with counts:
Follow your agent instructions in agents/auditor.md and reference the security-audit skill and all relevant checklists."
Stats
Stars1
Forks1
Last CommitNov 13, 2025