Multi-model adversarial code review with a deterministic, machine-computed release verdict. Portable agent skill for Claude Code, OpenAI Codex, and any SKILL.md-compatible platform.
This repository also configures its own agents. See what adversarial-review tells them →
Skill Codex
Independent adversarial review and deterministic release gating for code changes, using a multi-model reviewer panel over OpenRouter (or any OpenAI-compatible router, or an MCP transport like Composio). Use whenever the user asks for an adversarial review, red-team review, multi-model review, independent review…
Skill Claude CodeCodex
Verify what the code actually does against what the docs, roadmap, or handoff claim it does — the code is the source of truth. Use before relying on any documented behavior or status.
Skill Claude CodeCodex
Make the PR description, commit messages, and CHANGELOG match what actually changed — no drift, honest status, docs in the same change. Use before opening or updating a PR.
Skill Claude CodeCodex
Every claim ships with the artifact that proves it — test output, a run dir, an attestation, a review verdict. Use when finishing a change or asserting that something works.
Skill Claude CodeCodex
Write the spec before the code, make the spec the contract, and loop until the implementation is verified against it. Use before starting any non-trivial change in this repo.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: