The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes project source code to extract API annotations, which is an inherent part of its function but represents a standard ingestion surface. The severity is considered safe as it is central to the skill's purpose.\n
Ingestion points: The skill reads Go source files (**/*.go) to parse Swagger comments.\n
Boundary markers: There are no explicit instructions to the agent regarding the separation of code comments from its operational logic.\n
Capability inventory: The skill possesses file writing and command execution capabilities necessary for documentation generation.\n
Sanitization: Extracted annotation content is processed by the standard swag utility without additional intermediate filtering.\n- [EXTERNAL_DOWNLOADS]: The skill automates the installation of documentation utilities from well-known sources during setup.\n
Evidence: It executes go install github.com/swaggo/swag/cmd/swag@latest, targeting a standard and widely-used tool in the Go ecosystem.\n- [COMMAND_EXECUTION]: The skill utilizes command-line tools to generate and maintain documentation.\n
Evidence: It calls swag init and swag fmt via a restricted shell environment to process API documentation metadata.

golang-swagger