skf-rename-skill
Pass
Audited by Gen Agent Trust Hub on May 11, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to perform filesystem operations such as copying, moving, and deleting directories, as well as searching file contents with grep. These operations are essential for the skill's primary purpose of renaming skill packages. The workflow incorporates significant safety measures, including input validation via regex, collision checks, and a transactional flow that ensures the new version is verified before the old version is deleted.
- [DATA_EXFILTRATION]: There are no network operations, API calls, or attempts to transmit data externally. The skill's activities are strictly confined to the local project directories defined in the configuration.
- [PROMPT_INJECTION]: The instructions are procedural and highly specific, focusing on file management. No patterns associated with prompt injection, role-play bypasses, or instruction overriding were detected.
- [EXTERNAL_DOWNLOADS]: The skill does not fetch external code, scripts, or packages. All logic is self-contained or refers to internal project files.
- [SAFE]: The skill author (armelhbobdad) uses standard naming conventions and follows a structured development workflow. The use of 'agentskills.io' as a reference point for naming conventions and registry warnings is appropriate for the skill's context.
Audit Metadata