darwinian-objectivity
Pass
Audited by Gen Agent Trust Hub on Apr 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains no executable code, scripts, or binary files. It consists entirely of markdown instructions and a JSON file for test prompts.
- [PROMPT_INJECTION]: No evidence of instructions designed to bypass safety filters, extract system prompts, or override agent behavior was found. The instructions focus on methodology for testing user hypotheses.
- [DATA_EXFILTRATION]: There are no network operations, hardcoded credentials, or access to sensitive local file paths (e.g., .ssh, .env).
- [REMOTE_CODE_EXECUTION]: The skill does not include any dependency management files (requirements.txt, package.json) or commands that download and execute remote scripts.
- [INDIRECT_PROMPT_INJECTION]: While the skill ingests user input in the form of opinions and hypotheses, it lacks 'capability inventory' such as shell execution, file writing, or network access that would allow an injection to be exploited. The risk is assessed as safe because the agent only performs textual analysis on the input.
Audit Metadata