The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's primary function is to generate recipes for the just command runner. It incorporates security best practices by recommending set shell := ["bash", "-euo", "pipefail", "-c"] to ensure scripts are resilient and fail immediately upon encountering errors or unset variables.
[DATA_EXPOSURE]: It provides templates for managing environment variables via set dotenv-load and set dotenv-required. The instructions include explicit warnings against hardcoding sensitive credentials in justfiles, advising the use of environment variables or .env files instead.
[INDIRECT_PROMPT_INJECTION]: The skill automates the creation of an AGENTS.md file or updates existing agent context files (e.g., .cursor/rules, CLAUDE.md). These updates provide explicit instructions to future agent sessions on how to correctly utilize the project's verification gate (just build). While this influences future agent behavior, it is a transparent and intended mechanism for maintaining consistent development workflows.
[EXTERNAL_DOWNLOADS]: The documentation and templates reference well-known development tools and official GitHub Actions, such as extractions/setup-just, jq, and watchexec. These references are provided for environment configuration and task automation using trusted sources.

justfile