The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes dynamic context injection (the ! syntax in SKILL.md) to execute shell commands like find and pwd at load time. This includes checking for the existence of .env files to determine if the Justfile should be configured with environment variable loading. While this probes for sensitive files, the check is functional for the skill's purpose and restricted to the project root.
[PROMPT_INJECTION]: The skill demonstrates an indirect prompt injection surface as it processes untrusted data from local project files to influence its output and configuration generation.
Ingestion points: Content is read from Justfile, Makefile, and project manifest files such as package.json, pyproject.toml, and Cargo.toml during the detection and analysis steps.
Boundary markers: The instructions do not define delimiters or provide warnings to the agent to treat the content from these project files as untrusted or to ignore instructions embedded within them.
Capability inventory: The skill has access to powerful tools including Bash, Write, and Edit, which could be manipulated if malicious content in the scanned files influences the agent's logic.
Sanitization: There are no mechanisms described for sanitizing or validating the inputs from external project manifest files before they are used to populate the compliance report or generate new configuration files.

configure-justfile