How to Run a Pre-Mortem Risk Analysis Using the /pre-mortem Command

The /pre-mortem command loads the pre-mortem skill definition from the pm-execution plugin and processes your product requirements to generate a categorized risk report identifying Tigers, Paper Tigers, and Elephants before launch.

The /pre-mortem command in the phuryn/pm-skills repository enables product teams to conduct proactive risk analysis by imagining a launch has already failed. As part of the pm-execution plugin, this command leverages a Markdown-based skill architecture to analyze your product requirements document (PRD), launch plan, or feature description without requiring custom code. This guide explains how to execute a pre-mortem risk analysis using the command structure defined in the pm-skills source code.

What Is the /pre-mortem Command?

The /pre-mortem command is a thin wrapper defined in pm-execution/commands/pre-mortem.md that interfaces with the pm-toolkit runtime. According to the phuryn/pm-skills source code, the command parses its header block to expect a free-form argument (<PRD, plan, or feature description>). When invoked, the runtime automatically maps this command to the corresponding skill definition located at pm-execution/skills/pre-mortem/SKILL.md, injecting the $ARGUMENTS placeholder to pass your input to the AI.

How the Pre-Mortem Workflow Executes

1. Command Dispatch and Argument Parsing

The command file pm-execution/commands/pre-mortem.md contains a YAML header block (delimited by ---) that instructs the pm-toolkit runtime to expect a free-form text argument. This argument accepts raw text descriptions or uploaded documents via the <upload: filename> syntax.

2. Skill Loading and Execution

When triggered, the runtime searches for a skill matching the command name (pre-mortem) and loads pm-execution/skills/pre-mortem/SKILL.md. The skill's Instructions section directs the AI (Claude) to simulate a failed launch scenario, systematically enumerating potential risks from your input material.

3. Risk Classification and Ranking

The skill categorizes identified risks into three distinct types:

• Tigers: Real, launch-threatening risks requiring immediate mitigation
• Paper Tigers: Overblown concerns that appear dangerous but pose minimal actual threat
• Elephants: Obvious, unaddressed issues that everyone sees but no one has documented

For Tigers, the system further ranks them by urgency into launch-blocking, fast-follow, and track categories.

4. Report Generation

The skill formats output as a structured Markdown report containing a risk table with columns for Likelihood, Impact, Mitigation strategy, Owner, and Deadline, plus a Go/No-Go checklist.

Running a Pre-Mortem Analysis

Basic Text Input

Invoke the command followed by your feature description:

/pre-mortem We're launching a self-serve billing portal next month with Stripe integration and user onboarding flows

Uploading a PRD Document

For comprehensive analysis of existing documentation:

/pre-mortem <upload: MyProduct_PRD.pdf>

The pm-toolkit's file-ingestion layer extracts relevant sections from PDFs or other document formats before processing.

Sample Output Structure

The generated report follows this exact structure:

## Pre-Mortem: Billing Portal

### Tigers (Real Risks)

| # | Risk | Likelihood | Impact | Mitigation | Owner | Deadline |

|---|------|------------|--------|------------|-------|----------|
| 1 | Payment gateway downtime | High | High | Add fallback provider | Engineering | 2026-07-01 |
| 2 | Inadequate onboarding | Medium | Medium | Create step-by-step guide | UX Lead | 2026-06-25 |

### Paper Tigers (Overblown Concerns)

- "Customers will hate the UI" – usability testing shows 90% satisfaction.

### Elephants in the Room

- Cross-team hand-off for tax compliance is not documented; schedule a sync.

### Go/No-Go Checklist

- [ ] All launch-blocking Tigers mitigated
- [ ] Fast-follow plan documented
- [ ] Monitoring in place for Track Tigers

Customizing the Analysis

Because both the command and skill are pure-Markdown specifications, you can modify behavior by editing pm-execution/skills/pre-mortem/SKILL.md. The underlying pm-toolkit automatically maps the command to the skill and returns the structured output, so no custom code compilation is required to adjust risk categories or output formats.

Summary

• The /pre-mortem command in phuryn/pm-skills wraps the pre-mortem skill to analyze product documents for potential launch failures.
• The workflow involves command dispatch in pm-execution/commands/pre-mortem.md, skill loading from pm-execution/skills/pre-mortem/SKILL.md, and AI-driven risk enumeration.
• Risks are classified as Tigers (real threats ranked by urgency), Paper Tigers (overblown concerns), and Elephants (unaddressed obvious issues).
• Input accepts free-form text or uploaded documents via the <upload:> syntax, with the runtime injecting content via the $ARGUMENTS placeholder.
• Output generates a structured Markdown report with mitigation tables and Go/No-Go checklists.

Frequently Asked Questions

What file types does the /pre-mortem command support?

The command accepts free-form text input directly or document uploads via the <upload: filename> syntax. According to the pm-toolkit architecture implemented in phuryn/pm-skills, the file-ingestion layer processes PDFs and other standard document formats to extract relevant sections for analysis.

Where is the pre-mortem logic actually defined?

The core logic resides in pm-execution/skills/pre-mortem/SKILL.md, which contains the Instructions section that drives the AI analysis. The command wrapper in pm-execution/commands/pre-mortem.md simply defines the interface and maps user input to this skill.

Can I modify the risk categories (Tigers, Paper Tigers, Elephants)?

Yes. Because the system uses Markdown-based skill definitions, you can edit pm-execution/skills/pre-mortem/SKILL.md to adjust the classification criteria, add new risk categories, or modify the ranking system for Tigers without writing custom code.

How does the command handle follow-up actions after the report?

The command definition optionally suggests next steps such as updating the PRD, creating test scenarios, or drafting a launch checklist based on the identified risks. These follow-up actions are defined in the skill specification and trigger based on the generated risk report output.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}