The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process potentially untrusted data from fiction manuscripts and project-related documents, and it possesses the capability to write changes back to the filesystem.
Ingestion points: The agent is instructed to read story-bible.md, session summaries in the sessions/ directory, and user-provided manuscript chapters to establish project context.
Boundary markers: The instructions do not define specific delimiters or "ignore embedded instructions" warnings for the content read from these files.
Capability inventory: The skill explicitly uses the str_replace tool to perform surgical edits on chapter files and documentation.
Sanitization: There are no explicit sanitization or validation mechanisms described for the text ingested from the story files.

fiction-workshop