The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses basic shell commands (ls, cat, grep) to inspect project files for the purpose of identifying the tech stack. These operations are read-only and targeted specifically at project manifests (e.g., Cargo.toml, go.mod), which is standard behavior for a documentation tool.
[DATA_EXPOSURE]: While the provided templates and examples mention sensitive concepts such as API_KEY, DATABASE_URL, and .env files, these are strictly placeholders for documentation purposes. The skill does not instruct the agent to access, store, or transmit the user's actual secrets or credentials.
[PROMPT_INJECTION]: The skill follows a structured workflow for generating Markdown content. It does not contain instructions that attempt to bypass safety filters or override the agent's core behavioral guidelines.
[EXTERNAL_DOWNLOADS]: The skill references several well-known and trusted services for badges and package management (e.g., Shields.io, GitHub, npmjs.com, PyPI). These references are used for generating valid links in documentation and do not involve executing untrusted remote code.

create-readme