The Agent Skills Directory

[SAFE] (SAFE): Analysis of the skill metadata and documentation found no evidence of prompt injection, obfuscation, or malicious instructions. All content aligns with the stated purpose of regulatory affairs management.
[Data Exposure & Exfiltration] (SAFE): The regulatory_tracker.py script manages data in a local JSON file (regulatory_submissions.json). No network operations (curl, requests) or access to sensitive system paths (e.g., SSH keys, credentials) were detected.
[Unverifiable Dependencies & Remote Code Execution] (SAFE): The Python script uses only standard libraries (json, datetime, typing, dataclasses, enum). There are no external package requirements or patterns suggesting remote code execution.
[Indirect Prompt Injection] (LOW): The skill possesses a data ingestion surface via the JSON tracking system. However, it lacks high-risk capabilities—such as command execution, dynamic evaluation, or network requests—that would be necessary to exploit instructions embedded in the ingested data.
[Dynamic Execution] (SAFE): No use of eval(), exec(), or unsafe deserialization (like pickle) was found. The script uses the safe json.load() method for data persistence.

regulatory-affairs-head