The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it reads repository content to detect project languages and executes linter tools (such as ESLint, Ruff, and golangci-lint) on the existing codebase during the verification step.
Ingestion points: Repository files and linter output during the verification step (SKILL.md).
Boundary markers: No specific delimiters or instructions to ignore embedded content within the source files or linter reports are provided in the instructions.
Capability inventory: The skill can write configuration files to the repository and execute shell commands via npm, pip, go, and git.
Sanitization: The skill does not perform sanitization of the repository content before processing or linting it.
[COMMAND_EXECUTION]: The skill uses subprocess execution to manage project dependencies, initialize git commits, and verify the configuration.
Evidence: The instructions in SKILL.md specify executing npm install, pip install, go install, git add, git commit, and linter commands such as npx eslint, ruff check, and golangci-lint run.
[EXTERNAL_DOWNLOADS]: The skill downloads and installs development tools from remote sources to set up the environment.
Evidence: Fetches the golangci-lint tool from its official repository using go install github.com/golangci/golangci-lint/cmd/golangci-lint@latest (SKILL.md).

code-style