dvy1987/agent-loom / .agents/skills/customer-discovery
customer-discovery
VerifiedRun Mom Test–style customer-discovery interviews to validate or kill an unbuilt idea — generate a non-leading interview guide, conduct or coach the conversations, and synthesize signal vs compliments. Load when the user asks to do customer discovery, run problem interviews, validate an idea with users, run a Mom Test, design an interview guide, or says "talk to customers", "validate the problem", "interview users", "Mom Test this", "did real users want it", "synthesize my interviews", "I just talked to N people". Sub-skill of `venture-exploration`. Hard-bans "would you use this?", solution-pitching, friend/family-only ICP, and treating compliments as validation. Calls `secure-*` before synthesizing any pasted external transcripts.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 9.9 KB |
| references/examples.md | markdown | 1.9 KB |
| references/interview-guide-template.md | markdown | 4.0 KB |
| references/mom-test-rules.md | markdown | 3.7 KB |
| references/synthesis-template.md | markdown | 4.3 KB |
references/mom-test-rules.md
View on GitHub# Mom Test Rules
Load in `customer-discovery` Step 4 / Step 5. Based on Rob Fitzpatrick's *The Mom Test* (2013) — the canonical rules for problem-discovery interviews that even your mum couldn't lie to you about.
## The 3 Core Rules
### Rule 1 — Talk about their life, not your idea
The interviewee must not know what you're building during the problem questions. The moment they know, every answer is shaped by polite agreement / disagreement.
Apply: lead with their day, their workflow, their pain. Save the solution reveal for after currency questions, if at all.
### Rule 2 — Ask about specifics in the past, not generics or opinions about the future
Past behaviour is data. Future-tense answers are noise. Forecasts of own behaviour are systematically over-optimistic.
Apply: never ask "would you", "could you", "do you think you would". Always ask "tell me about the last time", "walk me through", "what did you actually do".
### Rule 3 — Talk less and listen more
Founders dominate discovery interviews. Ratio target: founder talks 20%, interviewee 80%. Long pauses are gold — they trigger elaboration.
Apply: ask, then shut up. Repeat the last 3 words in upward intonation if they trail off.
## Bad questions → Good questions
| Bad | Good |
|---|---|
| Would you use a tool that automates X? | Tell me about the last time you did X. What did you actually do? |
| Do you think this is a good idea? | What's the worst part of doing X today? |
| Would you pay $30/mo for this? | Have you ever paid for something to solve this problem? What? |
| Is this a problem for you? | Walk me through the most recent time this came up. |
| How much would you pay? | What's it costing you today — in time, money, or sanity? |
| Would you recommend this to others? | Who else do you know who has this problem? |
| If we built X, would you switch? | What made you switch from <previous tool> to <current>? |
## The 5 Strong Signals (in order of weight)
1. **Currency** — they have spent money, time, or political capital on the problem. Highest weight.
2. **Workaround** — they have built / hired / cobbled-together a specific solution that's painful. High weight.
3. **Emotional intensity** — unprompted complaints, swearing, frustration. Medium-high weight.
4. **Past attempts** — they tried product X, Y, Z and abandoned them. Medium weight (tells you what NOT to be).
5. **Referrals** — they know 3+ other people with the same problem. Medium weight.
## The 3 Weak Signals (discard for verdict)
1. **Compliments** — "Cool idea", "I'd totally use that", "You should build this". Code as `COMPLIMENT`. Discard.
2. **Generic complaining** — "Yeah, X is annoying" with no specifics. Probe or discard.
3. **Hypothetical excitement** — enthusiasm AFTER the solution is revealed. Code as `SOLUTION-TINTED`. Discount heavily.
## Anti-Patterns During Interview
- **Pitching** — describing the solution before currency questions. Stop yourself.
- **Confirming hypothesis** — "So it sounds like X is your biggest issue?" → leading. Replace with "What's the biggest issue?"
- **Jumping to the next question** — silence is uncomfortable but productive. Wait 3 seconds before re-prompting.
- **Defending current product / direction** — discovery is not the moment to argue.
- **Taking compliments seriously** — it is the founder's job to discount them.
## When to Pitch
Optionally, in the wrap (last 3 minutes) AFTER currency questions, AFTER you've extracted the past-behaviour data. Mark anything said post-pitch as `SOLUTION-TINTED` in synthesis.
If the interviewee asks "what are you working on?" early — answer briefly and pivot back: "I'd love to tell you, but I want to learn from your experience first — it shapes what I build."
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the work | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation code | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |