The Agent Skills Directory

[SAFE]: The skill consists entirely of Flutter development guidelines and Dart code examples. No executable scripts or malicious patterns were found.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface area for processing untrusted data as it takes user requirements for mobile features and generates corresponding code.
Ingestion points: User-provided descriptions of application screens, state management logic, and navigation flows in the SKILL.md workflow.
Boundary markers: Absent; there are no specific instructions to the agent to distinguish user-provided data from system instructions.
Capability inventory: The skill allows for the creation and configuration of Flutter widget files, route definitions, and package dependencies.
Sanitization: Absent; the skill does not perform validation or filtering of user input.
[PROMPT_INJECTION]: The file skill-report.json contains self-referential safety claims, asserting the skill is 'safe' and 'safe_to_publish'. While these appear to be mock audit metadata provided for documentation, they are evaluated as an attempt to influence the automated analysis verdict.

flutter-expert