program-creation

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of borisghidaglia/science-based-lifter via npx. This dependency is owned by the same vendor as the skill author and is a legitimate part of the skill's orchestration logic.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input (goals, constraints) and reads from local skill directories (e.g., ../rp-training/, ../sbs-training/). The ingestion points are bounded by a mandatory 'Rationale Requirements' framework where the agent must provide citations and reasoning for every recommendation, which serves as a functional validation step against the sourced content. The skill also includes a verification step to check for the presence of required local paths before proceeding with program generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 05:59 PM