The Agent Skills Directory

[SAFE]: The skill primarily consists of technical documentation and best practices for Go development, covering topics such as concurrency, error handling, and performance optimization.
[COMMAND_EXECUTION]: The skill uses a local Bash script (scripts/go-stack.sh) to discover project metadata and instructs the agent to run standard development tools like gofmt, go vet, golangci-lint, and project-specific task runners (e.g., make, task). These operations are within the normal functional scope of a development assistant and do not involve remote code execution or suspicious commands.
[INDIRECT_PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection as it processes untrusted data from the user's project files.
Ingestion points: The agent reads project-specific files including go.mod and various .go source files to identify the project stack and review code logic.
Boundary markers: Absent. The instructions do not specify delimiters or warnings for the agent when processing these external project files.
Capability inventory: The skill allows the execution of shell commands through the Bash tool for project discovery and validation. The agent is also expected to write or refactor code in the project directory.
Sanitization: Absent. No explicit sanitization or filtering of project file content is mentioned before it is processed by the agent.

pflow-golang