skills/product-on-purpose/thinking-framework-skills/think-speculative-harms-anti-goals/Gen Agent Trust Hub
think-speculative-harms-anti-goals
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, command executions, or network operations were detected. The skill is entirely composed of instructional markdown and templates used to guide an AI agent through a speculative design ethics exercise.
- [NO_CODE]: The skill does not include any executable scripts, binaries, or configuration files that could be used for malicious purposes. It functions as a structured prompt extension.
- [INDIRECT_PROMPT_INJECTION]: The skill provides an attack surface by ingesting user-supplied descriptions of product features to analyze. However, the instructions mandate a highly structured 'anti-goals register' output and include specific constraints on the narrative style (near-future, plausible), which mitigates the risk of the AI following instructions embedded within the user-provided design descriptions.
Audit Metadata