The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes git remote -v and inspects .git/config to identify the repository host and remote URLs. This is standard functionality for a tool designed to scaffold repository-specific configurations.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data by reading existing repository files such as package.json, CLAUDE.md, and CONTEXT.md to determine the project state and existing conventions.
Ingestion points: Reads files from the root and docs/ directories of the local repository.
Boundary markers: None explicitly defined in the instructions, though it relies on standard markdown headers for updates.
Capability inventory: Performs file writes to the local filesystem and executes local git commands.
Sanitization: The skill summarizes existing state rather than executing content found within these files.
[DATA_EXPOSURE]: The skill accesses repository metadata including remote URLs and local configuration files. This access is scoped to the current repository and is necessary for its stated purpose of configuring issue trackers and domain documentation.

setup