The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface as it ingests untrusted data from the local repository and user-supplied arguments.
Ingestion points: Reads user input via $ARGUMENTS (SKILL.md) and project files, specifically .env or configuration files, using Read, Grep, and Glob tools.
Boundary markers: None identified. The instructions do not specify delimiters or warnings to ignore instructions embedded within the files it reads.
Capability inventory: The skill has access to powerful tools including Bash, Write, and Edit, which allow it to execute shell commands and modify the file system based on the processed data.
Sanitization: There are no explicit instructions for sanitizing or validating user input or file content before it is used to generate code or execute shell commands.

newebpay-checkout