agently
Pass
Audited by Gen Agent Trust Hub on Jul 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides architectural and implementation guidance for the Agently AI framework. Detailed analysis of the instructions and reference files confirms that the behavior is consistent with a legitimate development tool and no malicious patterns were identified.
- [DATA_EXFILTRATION]: The skill documents a feature for reporting framework issues to the author's GitHub repository. This process is secured by explicit instructions requiring human-in-the-loop confirmation and mandatory sanitization of local context, including the removal of absolute paths, usernames, and authentication tokens before submission.
- [CREDENTIALS_UNSAFE]: The skill promotes secure secret management by instructing developers to use environment variable placeholders (e.g.,
${ENV.xxx}) in configuration files instead of hardcoding API keys or other credentials. - [COMMAND_EXECUTION]: Python code snippets provided in the documentation demonstrate standard framework API usage for defining agents and processing model requests. These examples do not involve executing untrusted remote code or dangerous shell commands.
Audit Metadata