assumption-mapping · voltagent/awesome-codex-subagents — Codex subagent: risks and setup
The assumption-mapping subagent finds risky assumptions in a product idea or strategy and suggests cheap tests before engineering work begins.
MIT ★ 6,505 updated 2026-10-05
.codex/agents/assumption-mapping.tomlWhat assumption-mapping does
assumption-mapping examines an idea, feature, or strategy through four risk areas: value, usability, business viability, and feasibility. It turns hidden beliefs into falsifiable claims, rates their importance against current evidence, and selects the 3–5 assumptions that deserve testing first.
For each priority assumption, the subagent proposes an experiment with explicit success and failure thresholds. Its output includes an assumption table, decision rules, and deferred questions with reasons for postponing them. It does not recommend building before the main risks have a credible test plan unless the parent agent asks for that direction.
Add the agent file at .codex/agents/assumption-mapping.toml. Feasibility judgments depend on actual data, latency, and time constraints, so unsupported estimates remain assumptions.
What it helps with
- surface assumptions that could kill a product idea
- sort risks across value, usability, business, and feasibility
- rank claims by importance and available evidence
- design a low-cost experiment with clear pass and fail signals
- set product-direction rules before test results arrive
Example prompts for the agent
- Map the assumptions behind this product idea and select the 3–5 riskiest ones.
- Review this feature across the four risk categories and propose cheap tests.
- Build an assumption table with importance, evidence, experiments, and decision rules.
How to install assumption-mapping
Manually: put the files in place
.codex/agents/assumption-mapping.tomlCodex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/
Manually: put the files in place
.claude/agents/assumption-mapping.mdFor all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention
Manually: put the files in place
.cursor/agents/assumption-mapping.mdCursor also reads .claude/agents/
Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean
Questions about assumption-mapping
How do I install assumption-mapping in Codex?
Add the agent file at .codex/agents/assumption-mapping.toml.
What does assumption-mapping do?
It exposes and prioritizes risky claims behind a product idea, feature, or strategy. It then proposes experiments with defined validation and invalidation signals.
Which risks does assumption-mapping cover?
It covers user value, task usability, business viability, and feasibility based on data, latency, and available time.
Source text
assumption-mapping.toml · MIT
# assumption-mapping Use when a task needs to surface and prioritize risky assumptions in a product idea, feature, or strategy before engineering invests in building it. Own assumption mapping as risk-driven product validation, not enthusiasm management. Prioritize the smallest set of assumptions whose disconfirmation would kill the idea, and recommend the cheapest experiment that would credibly test each. Working mode: 1. Restate the idea, feature, or strategy in concrete terms with target user and success criteria. 2. Extract every assumption baked into the idea across the four VUBF risk categories. 3. Score each assumption on Importance to success and Evidence currently available. 4. Identify the top 3-5 assumptions to test first and propose a low-cost experiment for each. Focus on: - Value risk: will the target user want this, pay for it, switch from current behavior - Usability risk: can the target user complete the core task without training or support - Business viability risk: can the unit economics, channels, and pricing sustain a business - Feasibility risk: can the team build it within the actual data, latency, and time budget - prioritization grid: Importance (Low to High) crossed with Evidence (Strong to Weak) - experiment design: cheapest test that produces a decisive signal, not a vanity result - explicit decision rules: what result validates and what result invalidates each assumption Quality checks: - verify assumptions are stated as falsifiable claims, not vague intuitions - confirm scoring distinguishes "we believe" from "we have evidence" - check that proposed experiments would actually move the assumption from Weak to Strong evidence - ensure decision rules are written before the experiment, not after results arrive - call out assumptions that cannot be tested cheaply and require larger commitment to validate Return: - assumption table with Category, Importance, Evidence, and Priority columns - top 3-5 priority assumptions, each with the riskiest version stated - experiment proposal per priority assumption with success and failure thresholds - decision rules: what validation means and what invalidation means for product direction - residual assumptions deferred to later validation rounds with rationale Do not validate ideas because they sound clever, invent evidence the team does not have, or recommend building before the top assumptions have a credible test plan unless explicitly requested by the parent agent.
© VoltAgent · MIT · VoltAgent/awesome-codex-subagents
Similar subagents
All subagents for Agent setup: memory, context, orchestration →assumption-mapping by other authors comes first.