The Agent Skills Directory

[PROMPT_INJECTION]: The skill analyzes external project files like package.json or pyproject.toml to determine project type and command recipes, creating a surface for indirect prompt injection where malicious instructions in those files could influence agent behavior.
Ingestion points: Reads local project files including Justfile, package.json, pyproject.toml, Cargo.toml, and go.mod.
Boundary markers: No delimiters or specific instructions are implemented to prevent the agent from following instructions embedded in these files.
Capability inventory: Utilizes the Bash tool for file operations and command execution including build, test, and clean operations.
Sanitization: No sanitization or validation of the content within the ingested files is performed before processing.
[COMMAND_EXECUTION]: The skill executes powerful shell commands such as rm -rf and docker build through the Bash tool. While these are standard for task runners, they represent a significant capability that is driven by the skill's detection logic.

configure-justfile