oxygn-cloud-ai
OrganizationDeep Adversarial Reviews — cross-model review skill for Claude Desktop & Claude Code CLI. Claude self-review meets GPT-5.6 cross-model verification.
Categories
Indexed Skills (2)
darc
Deep Adversarial Review for Claude Code CLI — iterative adversarial review using Claude self-review and OpenAI GPT-5.6 cross-model verification. Use when reviewing plans, specs, designs, or code changes for flaws before implementation. Accepts /DARc and /DARc:<rounds>. Requires OPENAI_API_KEY in environment.
dar
Deep Adversarial Reviews (DAR) — iterative adversarial review using Claude self-review and OpenAI GPT-5.6 cross-model verification. Use when reviewing plans, specs, designs, or code changes for flaws before implementation. Accepts /DAR:<rounds> where <rounds> is 0-9. Default is 3 rounds. /DAR:0 loops until zero CRITICAL/HIGH/MEDIUM findings. Requires OPENAI_API_KEY in environment. Use whenever the user asks for a review, audit, adversarial check, adversarial review, or DAR of any document or code.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.