algorithm-designer
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- PROMPT_INJECTION (SAFE): The content consists entirely of instructional templates for algorithm design. There are no attempts to override system prompts or bypass safety filters.
- DATA_EXFILTRATION (SAFE): No network operations (curl, wget, fetch) or access to sensitive file paths are present in the documentation.
- REMOTE_CODE_EXECUTION (SAFE): The skill does not download or execute any external scripts or packages. It is purely informational.
- COMMAND_EXECUTION (SAFE): There are no shell commands, subprocess calls, or administrative operations defined.
- CREDENTIALS_UNSAFE (SAFE): No hardcoded secrets, API keys, or private tokens were found in the metadata or body.
Audit Metadata