The Agent Skills Directory

[SAFE]: The skill does not include any executable scripts, binaries, or automated installation steps. It consists entirely of Markdown instructions, technical references, and evaluation criteria.
[SAFE]: All external URLs and package references point to well-known and reputable open-source Go libraries and official documentation. There are no signs of typosquatting or malicious remote code execution.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and audit Go source code files (**/*.go), which creates a theoretical surface for indirect prompt injection if those files contain malicious instructions disguised as comments or data. However, this is inherent to the skill's primary function and no exploitable vulnerability was found.
Ingestion points: Processes project files via **/*.go glob patterns.
Boundary markers: No specific delimiters or "ignore instructions" warnings are defined for the code ingestion process.
Capability inventory: The skill utilizes Read, Edit, Write, Bash, and Agent tools, providing a standard development environment capability set.
Sanitization: No explicit sanitization or filtering of input code is performed before processing.

golang-error-handling