The Agent Skills Directory

[COMMAND_EXECUTION]: The skill explicitly instructs the agent to execute shell commands using the bats and zunit testing frameworks to verify script behavior. This includes running specific test files, filtering by tags, and using subprocesses to test zsh-specific features.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze user-provided shell scripts and BATS test files. These files represent an attack surface where embedded instructions (e.g., in code comments) could potentially influence the agent's behavior.
Ingestion points: User-supplied shell scripts (.sh, .zsh) and BATS test files (.bats) referenced in the agent's context (e.g., SKILL.md).
Boundary markers: The skill does not define specific delimiters or instructions for the agent to ignore content inside the processed scripts.
Capability inventory: The skill enables shell command execution (bats, zsh, expect), file writing for creating stubs (cat > file), and PATH manipulation in SKILL.md and references/patterns.md.
Sanitization: No explicit sanitization or filtering of input script content is mentioned.
[EXTERNAL_DOWNLOADS]: The documentation references external resources and installation paths hosted on GitHub, including the vendor's own repository (synapseradio/ai-skills) and well-known open-source projects (bats-core/bats-core, zunit-zsh/zunit).

shell-testing