The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it is designed to ingest, analyze, and execute code from external Flutter projects which may be untrusted.
Ingestion points: The skill reads project source code, configuration files (such as pubspec.yaml), and test files from the directory provided in the path argument to tools like run_analyze and run_tests.
Boundary markers: There are no explicit instructions or delimiters provided to the agent to distinguish between the skill's operational instructions and potentially malicious instructions embedded within the project's data or comments.
Capability inventory: The skill can trigger network operations to download dependencies via run_pub_get and perform code execution via the run_tests workflow.
Sanitization: No sanitization, integrity checking, or sandboxing requirements are specified for the target project files before they are processed by the analysis and test tools.

flutter-qa-validator