network-watcher
Pass
Audited by Gen Agent Trust Hub on Mar 11, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is entirely instructional, providing a framework and checklist for an AI agent to identify malicious network patterns in other skills. It does not perform any automated actions or execute logic.
- [NO_CODE]: There are no executable scripts, binaries, or code dependencies included in this skill. The code snippets provided in the documentation are strictly for illustrative purposes to train the agent on what to look for during an audit.
- [SAFE]: The skill has minimal permissions, with network access, shell execution, and file writing explicitly disabled. The enabled
fileReadpermission is appropriate for its stated purpose of analyzing the configuration files of other skills.
Audit Metadata