The Agent Skills Directory

[SAFE]: The skill establishes a comprehensive SaaS development lifecycle that prioritizes security and quality through structured stages and mandatory review rounds.
[CREDENTIALS_UNSAFE]: Instructions correctly advise against placing secret keys in chat or code, mandating their use in gitignored .env files and providing guidance for secure key rotation and deletion after setup.
[INDIRECT_PROMPT_INJECTION]: The lifecycle includes mechanisms for the agent to review user feedback and activity to suggest improvements. While this creates an attack surface for indirect prompt injection, the skill includes explicit instructions for the agent to perform security audits and test the application for input-based vulnerabilities like XSS.
[EXTERNAL_DOWNLOADS]: The skill utilizes well-known services and author-owned tooling for standard operational tasks. All external dependencies and service integrations are aligned with the skill's primary purpose of application development and monitoring.

app-lifecycle