The Agent Skills Directory

[SAFE]: The skill focuses on legitimate performance engineering practices for the Go programming language, providing educational content and references to standard optimization techniques.
[EXTERNAL_DOWNLOADS]: The skill references several external packages and tools used for performance monitoring and benchmarking. All identified sources are either official Go project repositories, well-known technology organizations (e.g., Hashicorp, Elastic, Prometheus, Grafana, Uber), or the author's own verified repositories.
References golang.org/x/perf/cmd/benchstat for benchmark analysis.
References performance-oriented libraries such as github.com/prometheus/client_golang, github.com/grafana/pyroscope-go, and go.uber.org/automaxprocs.
[COMMAND_EXECUTION]: The instructions include standard shell commands for Go development workflows, such as running benchmarks (go test -bench), profiling (go tool pprof), and compiling with specific flags (go build -gcflags). These are standard tools for the intended task.
[INDIRECT_PROMPT_INJECTION]: The skill contains an attack surface for indirect prompt injection as it processes external Go source code and profiling data. However, the skill explicitly includes a "Scope and authorization" section that instructs the agent to disregard instructions found in browsed pages, log files, or command output, which mitigates this risk.

golang-performance