foundations-behavioral-economics
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, code, or scripts were detected. The skill consists entirely of markdown templates, JSON data, and YAML configuration used for reference and knowledge retrieval.
- [EXTERNAL_DOWNLOADS]: The
data/sources.jsonand various reference files cite numerous external sources. All identified URLs point to reputable academic publishers (e.g., Nature, Science, JSTOR), educational institutions (e.g., Yale, MIT), and official regulatory or intergovernmental bodies (e.g., UK ICO, US FTC, OECD, EU Commission). No suspicious or obfuscated links were found. - [COMMAND_EXECUTION]: The skill does not contain any executable scripts, subprocess calls, or shell commands. The Dynamic Context Injection pattern (
!command) was not present in the SKILL.md file. - [PROMPT_INJECTION]: There are no instructions designed to bypass agent safety filters or override system prompts. The skill actively promotes safety by including mandatory "Misuse boundary" sections and ethical harm tests for every behavioral primitive described.
- [DATA_EXFILTRATION]: No network operations, credential harvesting patterns, or access to sensitive file paths (such as .ssh or .env) were identified.
Audit Metadata