The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from user inputs and existing files to create or update documentation, creating a potential for malicious instructions in that data to influence agent behavior.
Ingestion points: Processes user-provided story descriptions and existing markdown files located in stories/, backlog/, requirements/stories/, and epics/ folders.
Boundary markers: The instructions lack explicit delimiters (e.g., XML tags or triple quotes) or warnings for the agent to ignore instructions embedded within the user-provided content.
Capability inventory: The skill utilizes file search, file read, and file write capabilities within the local workspace.
Sanitization: No sanitization, validation, or escaping of the external content is performed before the agent processes or writes the data to the filesystem.

user-story