The Agent Skills Directory

[SAFE]: The skill is designed for creative writing assistance and does not contain any malicious code, external dependencies, or network-exfiltration patterns.
[PROMPT_INJECTION]: The skill possesses a surface area for indirect prompt injection as it processes untrusted user data (manuscript content). Malicious instructions embedded within a chapter could theoretically attempt to influence the behavior of the agents during review or editing processes.
Ingestion points: User-provided text files located in project directories such as chapters/, characters/, and world/.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the skill's logic.
Capability inventory: The skill is capable of reading and writing files and spawning parallel agents via the Task tool to process large manuscripts.
Sanitization: There is no mention of sanitization or filtering of processed manuscript text.

fiction