The Agent Skills Directory

[SAFE]: The skill provides standard instructional content for code review processes, specifically tailored for the Contentstack Dart SDK. No malicious patterns such as prompt injection, data exfiltration, or unauthorized command execution were detected.
[DATA_EXPOSURE_&_EXFILTRATION]: The instructions explicitly include a security check to prevent the accidental inclusion of API keys, tokens, or secrets in pull requests, demonstrating a security-positive design.
[COMMAND_EXECUTION]: The skill mentions running standard development tools like dart analyze and dart test. These are expected commands within a Dart development environment and do not pose a risk in this context.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external data (code diffs and repository files). While it does not include explicit boundary markers or instructions to ignore embedded prompts in the code being reviewed, this is standard behavior for code review agents and no specific vulnerabilities were identified in the implementation.

code-review