The Agent Skills Directory

[PROMPT_INJECTION]: The skill is intended to process and analyze external Go source code files (.go) and module files (go.mod). This creates an indirect prompt injection surface where malicious instructions could be embedded within the source code to influence the agent's behavior.
Ingestion points: Processes .go files and go.mod files as specified in the skill's description.
Boundary markers: No explicit delimiters or instructions to ignore embedded natural language commands within the code files are provided.
Capability inventory: The skill uses mcp__acp__Bash to execute commands such as go test, go build, and go fix across the project files.
Sanitization: There is no evidence of sanitization or validation of the code content before it is processed or analyzed.
[COMMAND_EXECUTION]: The skill includes bash command snippets for common Go development tasks, such as formatting (gofmt), running static analysis (go vet, golangci-lint), and executing tests (go test -race). These are standard operations for the intended use case.
[EXTERNAL_DOWNLOADS]: Mentions well-known and trusted Go community packages and tools, including pgx, easyjson, sqlc, and fgprof. These are standard technical dependencies within the Go ecosystem and do not represent a security risk.

golang