# Assumption Mapping

> Assumption Mapping

- Skill: `sethdford/assumption-mapping-2` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add sethdford/assumption-mapping-2`
- Raw SKILL.md: https://api.skillmd.com/api/skills/sethdford/assumption-mapping-2/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: sethdford (https://skillmd.com/u/sethdford)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/sethdford/assumption-mapping-2

---

# Assumption Mapping

Make the **invisible scaffolding** of a plan visible. Most failures trace to **untested assumptions**, not weak execution—prioritize **disproof** where stakes are highest.

## When to Use
- New products, major technical bets, GTM plans, partnerships, or whenever confidence outruns evidence.
- Before scaling spend on a plan whose success depends on a few unstated “of course” beliefs.

## Behaviors
List assumptions the plan **requires** to be true. **Categorize**: **market** (demand, willingness to pay), **technical** (feasibility, performance, safety), **resource** (people, budget, time), **timing** (why now, competitive window). For each: **confidence** (high/med/low), **evidence** (facts vs beliefs), and **cost if wrong** (delay, rework, reputation, capital). Sort by **risk × uncertainty**—test **high impact, low confidence** first with the **cheapest** experiments that can falsify. Revisit after each learning; **retire** or **rewrite** assumptions explicitly.

## Examples
**Example 1:** “Enterprises will self-serve” → market assumption; run 5 structured interviews plus a concierge pilot before building full onboarding.

**Example 2:** “Our model hits latency SLO on CPU” → technical assumption; benchmark on representative data before committing to architecture.