AndyShaman

AndyShaman / premortem

Public

Premortem skill for Claude Code — finds the concrete ways a plan could fail before commitment. Multi-agent silent scan, mitigation triplets, history snapshots, reverse-premortem. Klein 2007 + Kahneman outside view.

agentic-workflow anthropic claude-code claude-skill cognitive-bias
15
1
100% credibility
Found May 11, 2026 at 15 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A Claude Code skill that guides users through a structured premortem process to identify potential failure points in plans, launches, hires, or decisions, saving results as markdown files.

How It Works

1
🔍 Discover premortem tool

You hear about a helpful way to check your big plans, launches, or decisions for hidden problems before they happen.

2
📥 Add to your AI assistant

You easily add this tool to your AI workspace so it's ready whenever you need it.

3
💭 Share your plan

You tell your AI helper about your specific plan, like a product launch or new hire, using simple words like 'premortem this.'

4
❓ Fill in key details

Your AI asks only for what's needed, like who it's for, what success looks like, and when to check back.

5
🕵️ AI spots the weak spots

Quietly, your AI looks from different angles and lists 5-8 real problems that could sink your plan.

6
🗣️ Pick fixes together

For each issue, your AI suggests smart options, you choose, and it helps plan the first steps to fix or watch them.

📄 Get your saved report

You receive a clear file with all issues, top priorities, advice, and history to revisit later and see progress.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 15 to 15 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is premortem?

Premortem is a Python skill for Claude Code that runs premortem analysis (premortem meaning: imagining failure before commitment, per Klein 2007 and Kahneman) on your plans, launches, hires, or decisions. It finds concrete ways things could fail using a multi-agent silent scan from six angles, proposes mitigation triplets, runs bias checks, and includes reverse-premortem thinking to counter over-caution. You get Markdown reports in docs/premortem/ with history snapshots for re-runs, all triggered by phrases like "premortem this: shipping new pricing."

Why is it gaining traction?

Unlike generic LLM risk lists, it forces narrative failure modes for honest, specific holes—backed by outside view and Kahneman checklists—delivering 5-8 deduped issues in 10-15 minutes. Persistent files track hole status over time (same, worse, resolved, new), and CLI tools like stats or snapshot let you query plans without Claude. Devs hook on the reproducible session that persists beyond chats, blending premortem methode with code workflow.

Who should use this?

Founders or PMs before product launches, pricing changes, or hires—like "first marketing role at $60k"—needing premortem analyse on concrete commitments. Tech leads evaluating architecture shifts or partnerships with clear horizons (3 months post-launch). Avoid for vague ideas or irreversible calls; pairs with Claude Code users doing plan_io in projects.

Verdict

Early alpha with 15 stars and 1.0% credibility score—light tests, bilingual docs, but unproven at scale. Worth a spin if you're in Claude Code daily; install via git clone and install.sh, run once on a real plan to gauge value before betting big.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.