The Agent Skills Directory

[SAFE]: The skill facilitates semantic code intelligence for Go development by integrating the official Go language server, gopls. This enables features like go-to-definition, find references, and safe refactoring within a Go workspace.
[EXTERNAL_DOWNLOADS]: The skill installs gopls using go install golang.org/x/tools/gopls@latest. This download targets the official, trusted domain for the Go programming language toolchain and is standard practice for Go development.
[INDIRECT_PROMPT_INJECTION]: The skill ingests local Go source code (**/*.go), which represents an attack surface for indirect prompt injection if a file contains malicious comments or code designed to influence agent behavior through diagnostics or documentation summaries.
Ingestion points: Project source files are read from the local filesystem (SKILL.md, references/matrix.md).
Capability inventory: The skill executes go, gopls, git, and golangci-lint to analyze and manage code (SKILL.md, references/mcp.md).
Boundary markers: The skill relies on standard tool outputs and does not include explicit prompt delimiters for the processed code.
Sanitization: Data is processed via standard semantic analysis tools which generate structured responses for the agent.
[COMMAND_EXECUTION]: The skill leverages official development binaries (go, gopls, git, golangci-lint) to perform code intelligence tasks. These tools are restricted to their intended purposes within the Go build and analysis environment.

golang-gopls