The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from existing project files (Makefiles, Rakefiles, package.json scripts) to perform migrations. This creates an attack surface where malicious or deceptive instructions within those source files could influence the agent's migration logic or be propagated into the new justfile.
Ingestion points: Scans project directory for existing build tool markers (e.g., Makefile, scripts/, package.json) to translate targets into recipes.
Boundary markers: No explicit instructions or delimiters are used to ensure the agent ignores potentially malicious code embedded in the source files being migrated.
Capability inventory: The skill has the capability to write new files (justfile, AGENTS.md) and modify existing project documentation and CI workflows.
Sanitization: No specific sanitization or validation logic is defined for the content extracted from source files during migration.
[COMMAND_EXECUTION]: The generated templates for various ecosystems (Python, Go, Rust, etc.) contain shell commands for common development tasks such as testing, linting, and cleaning. This includes powerful commands like rm -rf in cleanup recipes and find ... | xargs patterns.
[EXTERNAL_DOWNLOADS]: The skill references and recommends using the official just documentation, installation via standard package managers (Homebrew, Cargo), and the extractions/setup-just GitHub Action. These are well-known and established resources within the development community.

justfile