link-agentic-grimoire-custom
Pass
Audited by Gen Agent Trust Hub on Jul 21, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill runs a local bash script to perform file management tasks such as directory creation, file movement, and symbolic link generation within the user's home directory. These actions are consistent with the stated purpose of managing agent profiles and lack any malicious intent.
- [CREDENTIALS_UNSAFE]: The script includes a robust 'DENY' list that explicitly prevents the modification or linking of sensitive files containing account identifiers, authentication tokens, and session history (e.g.,
.claude.json,.credentials.json,settings.json), ensuring account isolation is maintained. - [DATA_EXFILTRATION]: No network activity or external data transmission was detected. The script's operations are confined to the local filesystem and do not interact with any remote endpoints.
Audit Metadata