The Agent Skills Directory

[PROMPT_INJECTION]: No attempts to override agent behavior, bypass safety guidelines, or extract system prompts were detected. The instructions are strictly focused on the technical workflow for Flutter plugin development.
[DATA_EXFILTRATION]: No commands accessing sensitive files (e.g., credentials, SSH keys) or performing unauthorized network operations were found. The use of 'flutter build' and 'flutter create' is standard for the domain.
[REMOTE_CODE_EXECUTION]: There are no patterns involving the download and execution of untrusted scripts (e.g., piping curl to bash). Commands are limited to official Flutter CLI tooling.
[COMMAND_EXECUTION]: The shell commands provided ('flutter create', 'flutter build') are standard for the development workflow described and do not use user-controlled input in a dangerous way.
[CREDENTIALS_UNSAFE]: No hardcoded secrets, API keys, or private tokens were identified. Organization placeholders (e.g., 'com.example.organization') are used correctly as examples.
[OBFUSCATION]: The content is presented in plain text without the use of Base64 encoding, zero-width characters, homoglyphs, or other techniques designed to hide malicious intent.
[PRIVILEGE_ESCALATION]: The skill does not request elevated permissions (e.g., 'sudo') or attempt to modify system-level configurations or file permissions unnecessarily.
[PERSISTENCE]: No mechanisms for establishing persistence, such as modifying shell profiles, cron jobs, or startup services, were detected.
[DYNAMIC_EXECUTION]: While the skill involves generating code via templates, it does not involve runtime compilation of untrusted input or the use of unsafe deserialization methods.

flutter-building-plugins