The Agent Skills Directory

[SAFE]: The skill operates in a read-only capacity, fetching metadata, transcripts, and agent revisions via the Duvo public API tools (e.g., getRun, listRunMessages, getRevision).
[SAFE]: All external domains, such as duvo.ai and github.com/duvoai/skills, and packages like @duvoai/cli, are official vendor resources belonging to the skill author 'duvoai'.
[SAFE]: No obfuscation, hardcoded credentials, or persistence mechanisms were detected in the instructions.
[SAFE]: The skill handles indirect data ingestion (transcripts and AOPs) for analysis only; it does not perform high-privilege actions like file writing or shell command execution. It explicitly delegates AOP rewrites to a separate aop-writer skill rather than performing them itself.

run-debugger