The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill operates on user-provided project files (source code, configuration files like pubspec.yaml, analysis_options.yaml) and utilizes powerful CLI tools such as dart analyze, dart fix, and dart test. This constitutes an inherent attack surface where malicious code or configurations within a processed repository could attempt to influence the agent's behavior.
Ingestion points: The skill reads project-level configuration and source code across the directory tree (e.g., lib/, test/, bin/).
Boundary markers: The instructions do not define specific delimiters or warnings for the agent to ignore instructions embedded in the analyzed data.
Capability inventory: The skill utilizes the full suite of Dart SDK tools for code analysis, fixing, testing, and compilation.
Sanitization: No explicit sanitization or filtering steps for the ingested data are defined.
[EXTERNAL_DOWNLOADS]: The FFI.md file provides a template for build hooks that download native binaries from GitHub releases. Crucially, the provided code includes a security best practice: verifying the downloaded file's MD5 hash against a local lookup table before registering the asset.
[COMMAND_EXECUTION]: The skill guides the agent to use standard Dart SDK command-line tools for project maintenance and validation, such as dart analyze, dart fix, dart format, and dart test.

dart