The Agent Skills Directory

[SAFE]: No malicious patterns or security risks detected.
[PROMPT_INJECTION]: The skill uses instructions like "Prefer retrieval-led reasoning over pre-training-led reasoning" to ensure the AI uses the provided documentation over general knowledge. These are benign instructions intended to improve accuracy and do not attempt to override safety filters or agent personality.
[EXTERNAL_DOWNLOADS]: The documentation contains setup instructions for developers, including curl commands to download assets and git commands to clone starter kits. All URLs point to the official vendor domain (img.ly) or well-known services like GitHub and StackBlitz.
[NO_CODE]: The skill consists entirely of markdown documentation and YAML metadata. There are no scripts or binary files included that execute during the skill's operation, minimizing the attack surface.

docs-angular