weekly-report-to-annual
Pass
Audited by Gen Agent Trust Hub on May 18, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes safe subprocess calls in its demonstration script to execute local Python scripts. It avoids shell invocation and uses absolute paths or script names with the current Python interpreter, minimizing the risk of command injection.
- [EXTERNAL_DOWNLOADS]: The script
fetch_emails.pyinitiates a network connection to the well-known Feishu IMAP server (imap.feishu.cn). This is a documented and necessary function for retrieving the emails required for the report. - [DATA_EXFILTRATION]: The skill handles sensitive user credentials for email access. Analysis confirms these are only used for authentication with the official service provider and are not exfiltrated to unauthorized destinations.
- [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes untrusted email content from an external source.
- Ingestion points: Email message bodies are fetched via the IMAP tool and read into the agent's working memory.
- Boundary markers: Delimiters such as '--- 周报数据 ---' are used in the demonstration logic to help the model distinguish data from instructions.
- Capability inventory: The skill possesses the ability to write files to the local disk via
save_report.py. - Sanitization: There is no automated sanitization of the email content before it is passed to the language model.
Audit Metadata