skills/kangarooking/poor-charlies-almanack-skill/misjudgment-psychology-checklist/Gen Agent Trust Hub
misjudgment-psychology-checklist
Pass
Audited by Gen Agent Trust Hub on Apr 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is purely educational and its behavior matches its description. No malicious patterns were identified.
- [NO_CODE]: The skill consists entirely of Markdown instructions and JSON test data; no executable scripts or system commands are present.
- [PROMPT_INJECTION]: No override instructions, safety bypasses, or system prompt extraction attempts were found in SKILL.md or test-prompts.json.
- [DATA_EXPOSURE_AND_EXFILTRATION]: There is no access to sensitive file paths (e.g., .ssh, .aws) and no network exfiltration patterns.
- [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill does not install third-party packages or download/execute remote code.
- [INDIRECT_PROMPT_INJECTION]: 1. Ingestion points: User decision scenarios in text format. 2. Boundary markers: None. 3. Capability inventory: No shell access, network operations, or file-writing. 4. Sanitization: Not applicable. The attack surface is non-exploitable due to zero available capabilities.
Audit Metadata