The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The installation process involves cloning a repository from an unverified GitHub account (retrocenterspark24) which is not a recognized trusted source. This repository contains the core logic for the SEO commands and multi-step workflows.
[COMMAND_EXECUTION]: The skill's documentation provides multiple examples of bash, Python, and Node.js automation scripts that utilize subprocesses to execute external commands. These scripts interact with system tools and network-based SEO services, posing a risk if the cloned code is not audited.
[CREDENTIALS_UNSAFE]: The configuration section instructs users to export sensitive API keys for external services (SEMRush, Ahrefs, Screaming Frog) into environment variables. These keys are then intended to be used by the scripts in the unverified repository, creating an exposure risk for credentials.
[PROMPT_INJECTION]: The skill possesses a significant ingestion surface as it is designed to analyze third-party websites for audits, competitor analysis, and ranking monitoring.
Ingestion points: Tools such as /technical-seo <domain>, /content-audit, and /competitor-gap fetch and process data from arbitrary external URLs.
Boundary markers: There are no specified boundary markers or instructions to the agent to treat external content as untrusted data.
Capability inventory: The skill uses network operations and shell command execution to perform its analysis tasks.
Sanitization: There is no evidence of sanitization or validation of the data retrieved from external domains before it is processed by the agent.

seo-content-marketing-command-suite