The Agent Skills Directory

[SAFE]: The skill consists exclusively of markdown documentation and rubrics. No scripts, binaries, or executable code are present in any of the skill files.
[SAFE]: The instructions emphasize a read-only approach and explicitly direct the agent not to install tools, start servers, or mutate the environment without user authorization.
[SAFE]: No malicious patterns, such as prompt injection, obfuscation, or data exfiltration, were identified.
[SAFE]: The skill defines an ingestion surface for external data (URLs, screenshots, and source code) but lacks associated tools or code that could be exploited for malicious execution or network exfiltration. The instructions further include boundary markers by advising manual verification of all findings.

ui-review