The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided narrative content ("Writer" inputs) to diagnose story structure problems. This creates a vulnerability surface where malicious instructions embedded within the analyzed story text could attempt to influence the agent's behavior or output formatting.
Ingestion points: User-provided descriptions of narrative choices, branching paths, and story states.
Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within the analyzed narrative data.
Capability inventory: The skill performs file system write operations to store diagnostic summaries (e.g., {project}-if-{date}.md) and configuration data (context/output-config.md or .interactive-fiction-output.md).
Sanitization: No explicit sanitization or validation of the ingested narrative data is described before it is processed or written to files.
[COMMAND_EXECUTION]: The skill interacts with the file system to save persistent diagnostic data and configuration. While these are documented functional requirements for the skill's purpose, they represent a capability that would be leveraged in the event of a successful prompt injection.
File writing: Instructions to write diagnostic analysis and user preferences to local files using specific naming patterns.

interactive-fiction