The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses standard shell commands like cat and ls to read project configuration and source files (README.md, package.json, pyproject.toml). These operations are restricted to the local project environment and are used solely to gather context for the rewriting task.
[DATA_EXFILTRATION]: No network-related commands or external API calls were detected. The skill processes information entirely within the agent's local context and does not attempt to send data to remote servers.
[PROMPT_INJECTION]: The instructions are focused on text transformation and style enforcement. There are no attempts to bypass safety filters, extract system prompts, or override agent constraints.
[SAFE]: The skill does not use any obfuscation techniques, does not attempt to gain administrative privileges, and has no persistence mechanisms. The metadata and referenced files are consistent with the stated purpose of humanizing documentation.

humanize-readme