pre-mortem

Devil's advocate for coding agents. Stress-tests plans, architecture and decisions with a steel-man, an assumption audit and a pre-mortem, then gives severity-rated findings and a verdict. Works in Claude Code, Codex, Cursor, Gemini CLI, Copilot and 12 more.

Devil's advocate skill/plugin for Claude Code and Cowork: turns Claude against you to harshly critique an idea or whole project across 8 angles and return a blunt verdict. Bilingual ES/EN. Made by tododeia.

A 9-step pre-mortem skill that interrogates you before you build. Grounded in 《置身钉内》(DingTalk ONE post-mortem). Works with Claude Code, Hermes Agent, and any markdown-skill agent.

Decision-quality pre-commitment analysis. Hosted at azimuth-testbed.netlify.app — no install required. Claude Code skill for permanent workflow integration.

A Claude plugin that turns scattered strategy artifacts into a testable, executable system — lifecycle, drift detection, and KPI coaching in one.

red-team-challenge: adversarial fact-check / red-team skill for Claude Code (with upgrade dogfood)

AI 사업 카운슬러 — Claude Code 플러그인. 아이디어를 13명 다관점·적대 토론으로 냉철 판독하거나 내 프로필 기반 AI 추천도 받고, 「한눈 요약」 카드로 즉시 확인 (로컬 저장·외부 전송 0)

A five-member expert council that pressure-tests consequential decisions for Claude and Codex.

Canary — hostile red-team failure analysis for software systems. Pre-mortem your codebase, find failure risks before they find you.

Future-you interrogates today's architecture decision, then files the regret before you commit it.

Convene a panel of expert archetypes to pressure-test any decision — get a clear PROCEED / CONCERN / BLOCK verdict with reasons. An open-source Claude skill from AOS-1.

Speed is the product. 11 Claude Code skills that help AI builders move fast, decide sharp, and 10x their impact.

AI-assisted pre-implementation reconnaissance for software changes — methodology, template, and case studies.

Skill for running a pre-mortem on a concrete plan before its next irreversible action. Surfaces decision-critical failure paths with evidence status, control assessment, residual-risk classification (Tiger/Open/Paper), and readiness verdicts. Not for vague intentions or retrospective root-cause analysis.

An agent skill that disagrees with you, on purpose, with evidence. Risk first, pre mortem always, verdict changes only on new evidence.

11-pass pipeline for reverse-engineering vague/lossy user prompts into structured analysis. Steel-manned restatement, abductive OARS extraction, first-principles decomposition, structural prior-art search, SCAMPER + constraint relaxation, pre-mortem, convergence, self-check, evidence verification, BLUF output. Portable to any agentic runtime.

Enterprise Cognitive Operating System, Empirical Architecture Blueprints & Closed-Loop Reasoning Standards for AI Coding Agents

Constrained multi-agent reasoning patterns — Six Thinking Hats, Pre-Mortem, Dialectic

Claude will tell you your idea is great. This skill makes it try to kill it first.

立项反问：在新项目 / 新合作 / 新标的启动前，用六问逼问把需求真实性、现状替代、最窄切口问到有证据为止。只出诊断与一个现实动作，不出方案。WorkBuddy & Claude Code 技能，纯 Markdown 无依赖。Adapted from gstack office-hours (MIT).