The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs repository operations using git and gh (GitHub CLI). It implements security best practices by using quoted heredocs ('EOF' and '__CE_PR_BODY_END__') when creating commit messages and PR bodies. This technique ensures that data processed by the agent is treated as literal text and cannot be executed as shell commands.
[DATA_EXFILTRATION]: The skill interacts with the GitHub API to manage pull requests. It includes specific defensive instructions to avoid using git add . or git add -A, which prevents the accidental staging and transmission of sensitive local files like .env, build artifacts, or credentials to remote repositories.
[EXTERNAL_DOWNLOADS]: The skill references assets from well-known services, specifically img.shields.io for status badges and github.com for repository links. These are standard, low-risk integrations for developer-oriented tools.
[SAFE]: No malicious patterns such as prompt injection, obfuscation, or persistence mechanisms were detected. The skill's logic focuses on legitimate developer productivity and includes proactive measures to maintain environment security.

ce-commit-push-pr