The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface through the ingestion of external data.
Ingestion points: The skill reads epic files and child story files (markdown with YAML frontmatter) from the project directory (e.g., in Step 2 and Step 4 of the workflow).
Boundary markers: The skill identifies document sections such as 'Epic Goal', 'Stories Breakdown', and 'Success Criteria' to scope its edits.
Capability inventory: The skill utilizes the Bash tool, Read tool, Edit tool, and shell execution (node -e) to perform its tasks.
Sanitization: The provided instructions do not explicitly describe sanitization or escaping mechanisms for content extracted from these files before they are processed or used in shell commands.
[DYNAMIC_EXECUTION]: The skill documentation in references/document-change-log.md describes the use of node -e to execute a dynamic JavaScript snippet for change log management. This script requires a local module (change-log.js) and processes command-line arguments to update markdown tables. While the script uses proper argument passing to avoid basic command injection into the JavaScript string, it represents a dynamic execution pattern for local file modification.
[COMMAND_EXECUTION]: The skill uses the Bash tool to list files and perform discovery logic in the file system. It also documents the execution of Node.js via the shell for administrative tasks related to document history.

edit-epic