The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes a local shell script (check.sh) to enforce house style rules and detect common AI-generated phrasing. The script uses standard system utilities like perl and grep to scan markdown files for banned characters (e.g., em dashes, curly quotes) and specific keywords. This is a legitimate use of shell execution for linting and quality assurance within a development workflow.
[EXTERNAL_DOWNLOADS]: The README.md file provides installation instructions using npx @tamng0905/builder-essential-skills. This package is associated with the skill's author and follows standard installation patterns for agent skills. No other external downloads or remote script executions were identified.
[DATA_EXFILTRATION]: No network operations or attempts to exfiltrate sensitive data were detected. The skill's primary operations involve reading project-specific content registries and writing blog posts to the local file system. References to external tools like /x-tweet are for intended distribution tasks and are part of the platform's standard capability set.
[PROMPT_INJECTION]: The skill does not contain instructions that attempt to override the AI agent's safety protocols or extract system prompts. The instructions are focused on adherence to the 'house voice' and technical writing standards.
[SAFE]: The skill operates on project-local data and follows established patterns for content creation and maintenance. The logic is transparent and the helper script performs benign analysis of prose content.

write-blog