The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted source content to generate platform-optimized versions, creating an attack surface where embedded instructions in the source text could influence agent behavior.
Ingestion points: Source content and metadata are collected in the initial workflow step defined in SKILL.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used when the agent processes the provided source content.
Capability inventory: The skill's core capability involves generating and writing markdown files to the local filesystem.
Sanitization: There is no evidence of sanitization or filtering of input content prior to processing.
[DATA_EXFILTRATION]: The included utility script scripts/multiplier.py performs file system operations using unvalidated user input.
Evidence: The --input argument allows the script to read any file accessible to the current user context.
Evidence: The --topic argument is used to construct output filenames ({topic}_{platform}.md) without sanitization, making the script vulnerable to path traversal that could write files outside the intended directory.

content-multiplier