feature-investment-advisor
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were detected in the skill contents.- [NO_CODE]: The skill consists exclusively of Markdown-based instructions and logic for the AI agent. It does not include, download, or execute any scripts, binaries, or external packages.- [PROMPT_INJECTION]: The instructions are clearly defined for financial analysis and do not contain any patterns intended to bypass safety filters or override system prompts.- [DATA_EXFILTRATION]: The skill lacks the capability to send data to external servers. No network-related commands (curl, wget, fetch) or non-whitelisted domains were identified.- [REMOTE_CODE_EXECUTION]: There are no patterns suggesting the download or execution of remote code. The skill relies on the agent's internal reasoning to perform basic arithmetic.- [DATA_EXPOSURE]: No hardcoded credentials, API keys, or access to sensitive file paths (e.g., .ssh, .aws) were found. All data processed is provided voluntarily by the user during the interactive session.- [INDIRECT_PROMPT_INJECTION]: While the skill ingests untrusted user data (business metrics), it possesses no exploitable capabilities such as file writing, network operations, or command execution, making the attack surface non-existent.
- Ingestion points: User-provided inputs in Step 0, 1, and 2.
- Boundary markers: None explicitly defined.
- Capability inventory: None (No code, no network, no file access).
- Sanitization: None required given the lack of dangerous capabilities.
Audit Metadata