tjhavranek

Mad Research — Security skill for Claude Code

Security community

Three Claude Code skills for working with Codex CLI: codex-bridge (one-shot Codex calls), mad-build (Claude+Codex collaboration with cross-review), and mad-research (three-stream adversarial audit of.

How to install Mad Research

This entry records only its repository, not the path inside it, so there is no exact command to give. Open tjhavranek/mad-research and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Mad Research does

Alternatives in Security

Skills — Trail of Bits Claude Code skills for security research, vulnerability detection, and audit workflows 4k ★
Anthropic Cybersecurity Skills — 734+ structured cybersecurity skills for AI agents · MITRE ATT&CK mapped · agentskills.io open standard · Work 3.8k ★
Amazon Skills — Free AI agent skills for Amazon sellers— keyword research, competitor analysis, listing audit & more 127 ★

README

mad-research: three Claude Code skills for working with Codex

[![DOI](https://zenodo.org/badge/1248112144.svg)](https://doi.org/10.5281/zenodo.20829175)

**For researchers who want a structured second opinion on a paper, proposal, or report, without writing code.**

Most [Claude Code](https://claude.com/claude-code) users who also use ChatGPT keep [Codex CLI](https://github.com/openai/codex) running in a separate terminal and shuttle outputs by hand. These skills let Claude call Codex directly from one session (for one-shot tasks, structured collaboration, or an adversarial audit) with no MCP server, no external orchestrator, no code to write.

Three independent skills in one repo. Install only the ones you want.

Skill	What it does	Use when...
`codex-bridge`	One-shot Codex call from inside Claude. No protocol.	You want Codex's view on a specific question without scaffolding.
`mad-build`	Four-step Claude+Codex collaboration: independent drafts → cross-review → revisions → merge.	You're producing something (code, draft, plan) and want a structured second pair of eyes.
`mad-research`	Three-role-stream adversarial audit of a document. Quote+page grounded criticism, anonymized cross-critique, fresh-context Codex synthesis against a locked rubric, minority report preserved. Optional opt-in Bayesian Mode (experimental: no documented end-to-end run yet) for evaluating the truth of a contested empirical claim.	You want a structured stress-test of a paper, grant, or referee report. See worked example.

These skills are the third, automated generation of the [Research Audit Duel + MAD protocols](https://github.com/tjhavranek/research-audit-duel-protocol) family (the manual protocols are versioned separately as Duel v1.7 / MAD v2.0: see Citation below), implemented for the Claude Code + Codex CLI combination.

**Looking for a deeper, Claude-only workshop?** The authors' successor s