The Agent Skills Directory

[SAFE]: The skill is purely instructional and provides safe coding patterns for Go development. No malicious behavior was identified across all threat categories.
[PROMPT_INJECTION]: The instructions are focused on providing technical guidance for Go testing and do not contain attempts to override safety guardrails or bypass system instructions.
[CREDENTIALS_UNSAFE]: No hardcoded API keys, tokens, or sensitive credentials were found in the skill metadata or provided examples.
[EXTERNAL_DOWNLOADS]: The skill does not perform network operations or attempt to download content from external domains.
[REMOTE_CODE_EXECUTION]: No remote script execution or untrusted code download patterns were detected.
[COMMAND_EXECUTION]: Usage of shell commands is strictly limited to the official Go toolchain via allowed-tools restrictions, which is necessary for the skill's stated purpose of testing and refactoring.
[OBFUSCATION]: The skill content and examples use clear, human-readable text and standard Go syntax. No hidden characters, homoglyphs, or encoded payloads were detected.
[INDIRECT_PROMPT_INJECTION]: The skill has a standard vulnerability surface for coding assistants. Ingestion points: The agent reads Go source and test files using Read and Glob tools. Boundary markers: None specified. Capability inventory: The skill utilizes Edit, Write, and restricted Bash (go, gofmt) tools. Sanitization: Standard agent processing with no additional input validation. The restricted toolset ensures this surface remains safe.

go-test-table-driven