The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes content from other skill files during audit and verification workflows. If an analyzed file contains malicious instructions, it could influence the agent.
Ingestion points: Processes SKILL.md and other files in workflows/ and references/ subdirectories (as seen in workflows/audit-skill.md and workflows/verify-skill.md).
Boundary markers: The skill does not use specific delimiters or instructions to ignore embedded commands within the files being audited.
Capability inventory: The skill possesses capabilities including directory creation (mkdir), file permission modification (chmod), and execution of various CLI tools.
Sanitization: External file content is read and analyzed without explicit sanitization or filtering logic.
[COMMAND_EXECUTION]: The skill executes shell commands to manage local files and verify the presence of CLI tools.
Evidence: Commands include ls, mkdir, chmod +x, which {tool}, and {tool} --version.

create-skill