The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes Go source code files (
Ingestion points: The agent reads project source files identified by the **/*.go pattern.
Boundary markers: The skill instructions do not explicitly require the use of delimiters or instructions to ignore embedded prompts when reading file content.
Capability inventory: The skill is configured with capabilities including Bash (for go, git, gopls, godig), Edit, Write, and Agent tools.
Sanitization: There are no explicit instructions for sanitizing or validating the content of the files before processing.
[COMMAND_EXECUTION]: The skill uses Bash to interact with standard Go development tools (go, gopls, golangci-lint) and version control (git). These are expected tools for a Go development persona and are correctly scoped to their respective binaries.
[DYNAMIC_EXECUTION]: The skill includes technical documentation on unsafe.Pointer and syscall.Syscall usage. This content is educational and provides specific safety rules (such as the six valid conversion shapes from the Go spec) to prevent common mistakes associated with low-level memory manipulation in Go.

golang-data-structures