migrate
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it reads and processes untrusted source files (.ts, .html) from a user-provided path.
- Ingestion points: Phase 1 involves reading all files in the target path or folder provided via $ARGUMENTS.
- Boundary markers: The instructions do not define clear delimiters or warnings to ignore instructions embedded within the source code being analyzed.
- Capability inventory: The skill has the capability to write to the local filesystem (applying migrations) and execute shell commands (Phase 3).
- Sanitization: There is no mention of sanitizing or escaping the content read from files before the agent processes them in the migration plan.
- [COMMAND_EXECUTION]: The skill executes several shell commands in Phase 3, including 'yarn format', 'nx run :build', 'nx run :lint', and 'nx run :test'. While these are standard tools in an Angular/Nx development environment, they represent an execution capability that could be targeted if the parameters (like library names or spec file paths) are manipulated via malicious source code during the analysis phase.
Audit Metadata