The Agent Skills Directory

[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection because it is designed to analyze untrusted external data such as source code and logs.
Ingestion points: Processes source code, server logs, pprof outputs, and web search results in files like SKILL.md, references/methodology.md, and references/testing-debug.md.
Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are defined for processing external code or logs.
Capability inventory: The agent can execute commands via Bash, launch sub-agents with Agent, and write files.
Sanitization: No explicit sanitization of input data is required before analysis or interpolation into prompts.
[COMMAND_EXECUTION]: The skill relies on the execution of various CLI tools for troubleshooting and diagnostics.
Evidence: Instructs the agent to use go build, go test, go vet, go mod, git, and the dlv (Delve) debugger through the Bash tool to diagnose issues and investigate the environment.
[EXTERNAL_DOWNLOADS]: The skill suggests installing external Go diagnostic utilities from public repositories.
Evidence: Recommends installing errcheck from github.com/kisielk/errcheck and the dlv debugger from github.com/go-delve/delve as part of the troubleshooting workflow. These are well-known community tools for Golang development.

golang-troubleshooting