The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute find and wc commands for counting Dart files and test files in the project. This is used solely for project scope discovery and progress reporting.
[DATA_EXPOSURE]: The skill is designed to read and analyze Dart source code files (**/*.dart) and test files (**/*_test.dart) within the user's project directory. This behavior is necessary for its primary function as a code auditor.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input in the form of third-party Dart source code.
Ingestion points: Reads all project Dart files using glob_file_search and the Read tool in references/architecture-compliance.md, references/code-standards.md, and references/testing-quality.md.
Boundary markers: Absent. The instructions do not specify using delimiters or explicit 'ignore embedded instructions' warnings when analyzing file content.
Capability inventory: The skill has access to Bash, Write, Edit, and WebFetch tools as defined in SKILL.md.
Sanitization: Absent. There is no evidence of sanitization or filtering of the code content before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: While the skill description mentions validating against 'live GitHub standards', the specific instructions in the reference files direct the agent to read standards from a local directory (agent-rules/rules/flutter/) within the somnio-ai-tools repository. The GitHub link provided in the metadata is informational.

flutter-best-practices