The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes basic shell commands using the Bash tool, specifically employing find and wc to discover and count Dart source files and test files within the project directory.
[EXTERNAL_DOWNLOADS]: The skill fetches the latest Flutter architecture and testing standards from the author's public GitHub repository (somnio-software/cursor-rules) using the WebFetch tool to ensure the audit is based on current best practices.
[PROMPT_INJECTION]: The skill analyzes external content (project source code and remote markdown files), creating a potential surface for indirect prompt injection where untrusted data could influence agent behavior.
Ingestion points: Local .dart and *_test.dart files, and remote .mdc standards files from GitHub.
Boundary markers: None explicitly implemented to delimit analyzed code from instructions.
Capability inventory: Tool access is restricted to reading files, basic shell discovery, and fetching remote standards.
Sanitization: No input validation or sanitization of the code files is performed before they are processed by the agent.

flutter-best-practices