The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified. The skill follows best practices for repository analysis and includes proactive security measures.
[DATA_EXPOSURE]: The skill performs local file system reads on project configuration files (e.g., package.json, Cargo.toml, .env) and existing documentation to determine the project type and content requirements. This access is necessary for the skill's functionality and is balanced by an automated PII/Infrastructure scrub that identifies and redacts sensitive data from the final output.
[COMMAND_EXECUTION]: Uses standard development CLI tools (git, gh, ls) to gather repository metadata, such as remote URLs and GitHub topics. These operations are restricted to the local repository context and serve the primary goal of README optimization.
[EXTERNAL_DOWNLOADS]: The skill installation instructions refer to the author's own GitHub repository (thatrebeccarae/claude-marketing). This is a standard and safe method for skill distribution by the author.
[PROMPT_INJECTION]: No evidence of instructions intended to bypass safety filters or override agent behavior was found. The instructions are focused strictly on the README generation and auditing lifecycle.

github-readme