The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands (cp, ls, and PowerShell's Copy-Item) to set up the project hierarchy and copy internal tool scripts from the skill package into the user's project. This is standard initialization behavior for a project scaffolding tool.
[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it performs a comprehensive scan of markdown files within the repository to evaluate migration status. This is a potential risk common to documentation-processing agents, though the skill incorporates human-in-the-loop mitigation by requiring user confirmation for all migration actions.
Ingestion points: Repository-wide scan for .md files defined in SKILL.md for auditing existing project state.
Boundary markers: None identified for distinguishing file content from instructions during the audit process.
Capability inventory: Shell command execution and file system modification (creating and updating .md and .yaml files).
Sanitization: No explicit sanitization or instructions to ignore embedded agent commands in source files are provided.

easysdd-onboarding