The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill maintains and reads from a local directory structure (~/community-manager/) to manage community data. Content stored in these files could theoretically be used for indirect injection, though the risk is low given the skill's restricted environment.
Ingestion points: Data is read from ~/community-manager/communities/{name}.md and other files within the memory structure.
Boundary markers: The skill does not define specific delimiters for separating external data from instructions.
Capability inventory: The skill performs local file read and write operations within its designated configuration path.
Sanitization: There is no evidence of sanitization for the content read from the community markdown files.

Community Manager