The Agent Skills Directory

[PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill is designed to ingest and process arbitrary text provided by users ("untrusted data") and possesses capabilities to modify files. * Ingestion points: Processes external text intended for humanization via the Read and Edit tools. * Boundary markers: The instructions do not specify the use of delimiters (such as XML tags or triple quotes) or provide warnings to the agent to ignore embedded instructions within the processed data. * Capability inventory: The skill utilizes sensitive tools including Write, Edit, and Glob, which could be misused if the agent obeys instructions hidden within the input text. * Sanitization: There is no evidence of validation or sanitization of the external content before it is interpolated into prompts or used in file operations.

humanizer