grow-app
Warn
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Operating Rule 4 instructs the agent to suggest a shell command (
npx skills add wondelai/skills/<slug> --global) to the user if a constituent skill is missing. This involves global software installation and execution of arbitrary code via the terminal. - [REMOTE_CODE_EXECUTION]: The fallback mechanism for missing skills relies on
npxto fetch and execute packages from a remote registry at runtime. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by reading and acting upon content stored in local project files (e.g.,
docs/GROW-APP-PLAN.md,PRODUCT.md). - Ingestion points: Files in the
docs/directory are read at the start of every session (Operating Rule 1). - Boundary markers: None; the skill does not include instructions to delimit or ignore potentially malicious content within these files.
- Capability inventory: The skill can perform shell command execution and filesystem writes.
- Sanitization: No validation or escaping is applied to the ingested documentation data.
Audit Metadata