The Agent Skills Directory

[SAFE]: The skill is designed for automated Go code review. It includes explicit security checks for SQL injection, command injection, path traversal, and hardcoded secrets.
[COMMAND_EXECUTION]: The skill utilizes Bash to run standard Go diagnostic tools such as govulncheck and go build -race. These operations are limited to the repository context and are standard for code quality assessment.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted code from repositories for analysis. This is the primary function of the skill. While it has shell access, the predefined commands are standard development tools. * Ingestion points: Reads files and git history via Read, Glob, Grep, and git diff. * Boundary markers: None explicitly defined in the instructions for the code input. * Capability inventory: Bash access for running Go build and test tools. * Sanitization: None specified for the content of the analyzed files.

go-backend-reviewer