The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the analysis and execution of external Go code, which introduces an indirect prompt injection attack surface. * Ingestion points: Source code from Go projects provided to the agent for review, refactoring, or modernization as described in SKILL.md. * Boundary markers: The instructions do not define specific delimiters or instructions to ignore potential commands embedded within the code files being analyzed. * Capability inventory: The skill enables the Bash(go:*) tool, allowing the agent to run potentially untrusted code through standard Go commands like go test, go vet, and go tool pprof. * Sanitization: The doctrine lacks instructions for sanitizing code inputs before they are processed by the agent or passed to the allowed shell tools.

golang-master