The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified. The skill is primarily composed of educational markdown files designed to improve the performance and safety of AI agents working with blockchain technology.
[EXTERNAL_DOWNLOADS]: The skill instructs agents to fetch current Ethereum development documentation from ethskills.com and GitHub repositories owned by austintgriffith. These references are intended to provide real-time updates on gas prices and contract standards, which are essential for the skill's stated purpose.
[PROMPT_INJECTION]: The contract auditing module (audit/SKILL.md) reads external, untrusted Solidity code as part of its systematic audit process. While this represents a surface for indirect prompt injection, the risk is inherent to the task of auditing and the skill provides defensive checklists and methodology to mitigate accidental obedience to embedded instructions.
[DATA_EXFILTRATION]: An automated scanner flagged a Chainlink data feed URL (data.chain.link) as potentially malicious. This is a false positive; Chainlink is a well-known technology provider and the link is used solely to provide factual price information to the agent.

ethskills