The Agent Skills Directory

[SAFE]: The skill serves as a style guide and procedural framework for technical writing within a code repository. It does not contain executable code or dangerous instructions.
[COMMAND_EXECUTION]: The skill instructs the agent to use standard development tools, such as gh repo view, to gather context about repository visibility and conventions. This is a benign use of established CLI tools for documentation purposes.
[REMOTE_CODE_EXECUTION]: In the verification step, the skill suggests running documented commands and examples to ensure they produce the expected output. This is a standard practice for technical documentation auditing and does not involve downloading or executing untrusted external scripts.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill explicitly includes a 'Stranger-safe' policy that directs the agent to identify and remove personal domains, account IDs, and private data from public-facing documentation, replacing them with fictional examples. This promotes data safety rather than compromising it.

repo-docs