Phase: Plan
Tags:#risk#planning#premortem
AI Model: Any (model-agnostic)
Use this prompt after creating an implementation plan to stress-test it before you start building. A premortem catches issues when they’re cheap to fix (in planning) rather than expensive to fix (in production).
| Variable | Description | Example |
|---|---|---|
| `` | The plan you’ve already created | (paste your plan) |
| `` | Relevant details about the existing system | “Next.js 14 monorepo, 50k LOC, 3 developers” |
| `` | How much time you have | “Must ship by end of sprint (5 days)” |
I've created the following implementation plan:
Codebase context:
Timeline:
Perform a premortem analysis. Assume the project has FAILED and
work backward to explain why. Cover:
1. **Technical risks** — What could break, conflict, or not work
as expected? Rate each as LOW / MEDIUM / HIGH.
2. **Dependency risks** — Are we relying on external services,
libraries, or APIs that could change or fail?
3. **Integration risks** — What existing code could this change
break? What side effects might occur?
4. **Timeline risks** — Is this plan realistic for the timeline?
What could cause delays?
5. **Knowledge gaps** — What do we not know that could derail us?
For each risk:
- Describe the risk
- Rate the likelihood (LOW / MEDIUM / HIGH)
- Rate the impact (LOW / MEDIUM / HIGH)
- Suggest a mitigation strategy
Present as a risk matrix table.
I've created the following implementation plan:
Plan: Add OAuth2 login (Google, GitHub) to our existing
email/password auth system using NextAuth.js v5.
Steps:
1. Install NextAuth.js v5
2. Configure Google and GitHub providers
3. Create auth API route
4. Update login page with social buttons
5. Migrate existing sessions
Codebase context: Next.js 14, Supabase Auth currently handles
email/password, 2000 active users
Timeline: Must ship by end of sprint (5 days)
[rest of prompt...]
A risk matrix identifying: session migration data loss (HIGH impact), NextAuth + Supabase auth conflict (MEDIUM likelihood), Google OAuth consent screen approval delay (HIGH likelihood for timeline), and mitigation strategies for each.