solidity-auditor

BENIGN for purpose-capability alignment, with moderate security risk. The skill’s behavior matches a Solidity auditing workflow, uses only a same-publisher GitHub VERSION check, and shows no credential harvesting or exfiltration path; however, it grants high-impact offensive security analysis capability and processes untrusted code through parallel agents, which raises medium overall risk.

The fragment is explicitly malicious in intent, providing a blueprint to exploit dependencies, token behaviors, and economic incentives in DeFi protocols. It is not a defensive code review or a safe, verifiable implementation. Treat as high-risk content; do not integrate; if found in a project, remove and implement defense-in-depth measures to prevent dependency manipulation, token misbehavior, and governance abuse.

This artifact is an explicit attacker playbook for exploiting access control weaknesses in smart contracts — its presence in a repository is a strong indicator of malicious intent (or at minimum adversarial guidance). It enumerates high-risk sinks (initialize, upgrades, grantRole, delegatecall, selfdestruct) and concrete exploit patterns (front-running initialization, guard-gap exploitation, confused deputy, storage collision). No benign code is present to analyze; therefore there are no false positives to defend. Because this is guidance rather than implementation, it cannot itself perform attacks, but it materially increases risk if used by an attacker against vulnerable contracts. Action: treat this as malicious guidance, remove from codebase, and audit any contracts in the project for the enumerated vulnerabilities (unguarded writers to shared storage, callable initializers on implementations, improper use of delegatecall/proxy storage layout, unchecked upgrade paths, and role management functions that accept address(0) or allow renounce without recovery).

This file is an explicit attacker playbook detailing practical execution-flow exploitation techniques against smart contracts and multi-transaction protocols. It contains actionable guidance on manipulating attacker-controlled inputs, encoding mismatches, stale reads, sentinel-value bypasses, mid-operation config mutation, and approval-residual abuse to reach high-value sinks (fund transfers, delegatecalls, state corruption). No runnable code is present, so I cannot trace concrete variables to a vulnerability in a specific contract; however, the guidance materially increases risk if associated with a repository or package. Treat presence of this content as a red flag: require immediate audit of nearby code, remove or flag the guidance, and investigate the authorship/context.