The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied application details that are interpolated into shell commands and a markdown report file.
Ingestion points: The APP_NAME, APP_TYPE, and TECH_STACK inputs collected in SKILL.md via environment variables or the read command.
Boundary markers: None present to isolate user-provided strings from the command structure.
Capability inventory: Shell command execution (bash), file creation/writing (echo), and text filtering (grep).
Sanitization: None. The script uses these variables directly in echo commands without escaping or validation, which could allow command substitution if malicious input is provided.

owasp-check