The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes user-provided Polish text files to remove AI-generated signatures, which creates an attack surface for indirect prompt injection where malicious instructions hidden in source files could influence agent behavior during the editing process.\n
Ingestion points: The skill reads and edits .md drafts and other local text files using the Read and Edit tools.\n
Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following commands found within the text files being processed.\n
Capability inventory: The skill is authorized to use Read, Write, Edit, Grep, and Glob tools, providing a path for file system modifications if an injection were to occur.\n
Sanitization: The skill lacks specific sanitization logic to filter out or neutralize instructions contained in the processed data.

humanizer-pl