The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands (including mkdir, cat, and date) to manage session directories and state files in a temporary workspace (found in resources/inspectional-reading.md and resources/component-extraction.md).- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection through its document processing workflow. It ingests untrusted content from a user-provided $SOURCE_DOC in resources/inspectional-reading.md and resources/component-extraction.md. Ingestion points: Content is read from $SOURCE_DOC for skim and component extraction steps. Boundary markers: Absent; there are no specific delimiters or instructions to ignore embedded commands within the source content. Capability inventory: The agent has the ability to execute shell commands and write files to the local file system. Sanitization: Absent; the content is processed for interpretation without prior validation or escaping.

skill-creator