munger-perspective
Pass
Audited by Gen Agent Trust Hub on May 26, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill utilizes role-play instructions to adopt the Charlie Munger persona ("Respond in Munger's identity", "Use 'I'"). These instructions are intended for the skill's primary analytical purpose and do not attempt to bypass core safety guardrails or extract system prompt information.
- [INDIRECT_PROMPT_INJECTION]: The skill's research-based workflow introduces a surface for indirect prompt injection by prioritizing external web search results for decision-making analysis.
- Ingestion points: External data retrieved via the WebSearch tool during Step 2 of the Answer Workflow.
- Boundary markers: Absent. There are no instructions in the skill to treat search tool outputs as untrusted data or to ignore embedded commands.
- Capability inventory: The skill utilizes web search capabilities. It does not define any file system write operations, subprocess executions, or direct network requests outside of standard tool use.
- Sanitization: Absent. The agent is encouraged to use raw findings from external research directly within its reasoning process.
- [SAFE]: No obfuscated content, hardcoded credentials, unauthorized dependencies, or persistence mechanisms were found. The skill does not contain executable scripts or shell commands.
Audit Metadata