The Agent Skills Directory

[SAFE]: No security issues were identified. The skill is composed entirely of informative Markdown documentation, workflow instructions, and project metadata. All identified patterns are consistent with the skill's stated purpose of providing security auditing guidance.
[EXTERNAL_DOWNLOADS]: The documentation references standard installation procedures for well-known and reputable security tools (e.g., Slither, Aderyn, Foundry, Echidna) from trusted package registries and official domains. These instructions are provided as reference material for the user/agent and do not represent autonomous or malicious execution of remote code.
[PROMPT_INJECTION]: The skill uses detailed instructional language to define its operational scope and workflow. No attempts to override agent safety protocols, role-play as a restricted entity, or extract system instructions were found. The extensive keyword list in the skill's description is a benign technique used to ensure the skill triggers appropriately for auditing-related queries.
[DATA_EXFILTRATION]: Analysis of the skill's instructions and reference materials found no evidence of hardcoded credentials, sensitive file path access (like SSH keys), or unauthorized network operations. The skill correctly promotes security best practices, such as using environment variables for configuration.
[COMMAND_EXECUTION]: The skill includes various shell command examples for using security scanners and utility scripts. These commands are benign, intended for code analysis, and are documented as part of the auditing workflow.

solidity-security-audit