reciprocation
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: No malicious patterns or security risks were identified during the analysis of the skill's instructions and documentation.
- [NO_CODE]: The skill is entirely documentation-based, providing methodology and case studies from a psychology book. No executable scripts, system commands, or external dependencies are included.
- [PROMPT_INJECTION]: Instructions focus on analyzing human social behavior. There are no attempts to bypass LLM safety guardrails, override system instructions, or extract system prompts.
- [DATA_EXFILTRATION]: No network connectivity, URL requests, or sensitive local file access patterns (such as credentials or environment variables) were detected.
- [OBFUSCATION]: The content is provided in plain text. No encoded strings, hidden characters, or deceptive homoglyph attacks are present.
Audit Metadata