The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions frequently use shell commands to operate the Reins CLI, specifically npx, bun, and cd commands across multiple workflow files.
[EXTERNAL_DOWNLOADS]: The skill relies on fetching and executing code from the npm registry at runtime using npx reins-cli@latest, npx skills check, and npx skills update. While these appear to be vendor-associated tools, they constitute runtime downloads of executable code.
[REMOTE_CODE_EXECUTION]: Instructions provide a path to execute local TypeScript code using bun src/index.ts <command> ../.. when working within the repository's own source structure.
[PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks as it is designed to read and analyze contents from arbitrary user repositories (e.g., AGENTS.md, ARCHITECTURE.md, and other documentation).
Ingestion points: The audit, doctor, and evolve workflows ingest data from a user-provided directory path.
Boundary markers: No explicit boundary markers or instructions to ignore embedded directives within the analyzed files are documented.
Capability inventory: The skill possesses the ability to execute shell commands (npx, bun) and write files to the local filesystem (e.g., during reins init or evolve --apply).
Sanitization: There is no evidence of sanitization or safety filtering of the data read from the target repository before it is processed by the agent.

Reins