layers-user-needs
Skill307 starsChanged 6 months agoMITView on GitHub
Techniques for eliciting and prioritising user needs, pains, and desires — the opportunities that feed product strategy
3 other projects use a version of this, 1 of them edited.
Adopted by
Most recently taken up by
- davidanthonyn/claude-skillsedited, first seen Oct 20, 2026
- pengGgxp/skillsword for word, first seen Oct 11, 2026
- miovrag/claude-skillsword for word, first seen Jun 12, 2026
Other versions3 projects use a version of this.
Word for word · 2
- miovrag/claude-skills★ 1100% same
- pengGgxp/skills★ 1100% same
Edited · 1
- davidanthonyn/claude-skills★ 020% same
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository8 other files it gives its agents.
- layers-conceptual-modelSkill
- layers-domainSkill
- layers-interaction-flowSkill
- layers-introSkill
- layers-observed-behaviourSkill
- layers-orientSkill
- layers-product-strategySkill
- layers-surfaceSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 23
If the needs are already clear and grounded, don't re-elicit them for the sake of it — take them to /layers-product-strategy.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/layers-user-needs
Assumes /layers-intro has been loaded. This skill is a library of techniques, not a script — see "How to use these skills" there.
User needs are what we think users are trying to achieve, and why — an interpretation built on observed behaviour and domain knowledge, not a direct capture of reality. This layer sits between the messy raw material of observation and the deliberate decisions of the solution space.
The outputs here are opportunities: needs (what users want to achieve), pains (what causes friction), and desires (improvements they'd value). All three are valid — elicit all three.
The decisions this layer makes
- Who exactly the users are whose needs we're defining — and in what situation
- What jobs they're trying to do: functional, emotional, and social
- Which needs are grounded in evidence, and which are assumptions
- Which needs matter most, and why
If the needs are already clear and grounded, don't re-elicit them for the sake of it — take them to /layers-product-strategy.
Disciplines — what keeps needs honest
- Need, not solution. "When I need a report, I want to export to CSV" is a solution. Push to the underlying need.
- Strip the mechanism. If the "When" clause references your specific solution (your dashboard, your CSM, your weekly email), you're describing the current system, not the need. Re-state it independent of who or what serves it: what's true about the user at the moment they need this? The need follows from the state, not the mechanism.
- The "When" must be picturable. Specific enough to see the moment it happens — triggered by an event, a feeling, a rhythm, or a threshold crossed. Push until it is.
- Elicit emotional and social jobs, not just functional. They're chronically under-articulated even when they're shaping behaviour. Asking explicitly is usually what surfaces them. (Functional → interaction & model; emotional → surface tone/feedback; social → surface, sometimes strategy.)
- Mark confidence: observed / inferred / assumed.
- Workarounds are signal. A need real enough to motivate a spreadsheet or a workaround email is a strong one.
Techniques
Job stories are the default; the rest suit particular situations.
| Technique | Use it when |
| Job stories (JTBD) | Default. When [situation], I want to [motivation], so I can [outcome]. Keeps solutions out; the "When" clause forces specificity. |
| User stories | The team prefers role-based framing or an existing Agile workflow. |
| Top tasks analysis (Gerry McGovern) | Large existing user base — identify which tasks matter most by frequency. Statistical, survey-based. |
| Persona + scenario | Communicating to stakeholders who think in archetypes. Good for alignment; less precise for design. |
| ODI desired outcomes (Ulwick) | Precise, measurable statements — "Minimize [metric] when [context]." Maps directly to opportunity scoring. |
| Surface hidden needs | Prompts to find what's ignored: what users do before/after the moment you focus on; what they wish they didn't have to do; what a workaround currently serves. |
| Rough prioritisation | Order by importance × how poorly currently served. A need that matters and is badly served is a high-value opportunity. Keep it rough — precise scoring is strategy work. |
Working with the designer
First settle who the users are and in what situation — not "users" but which type, when. If there's more than one distinct type with different needs, work them separately. Note the source (research, domain knowledge, or assumption); if it's assumption, mark it and plan to validate.
Then work the needs the designer raises through the disciplines above, and probe for hidden ones. Offer the technique that fits — job stories by default, ODI when measurability matters, top tasks when there's a large user base. Don't run a fixed sequence.
Capture only the residue: the prioritised needs with confidence ratings, the unprioritised-but-surfaced ones (so they aren't lost), gaps (probably-real needs not yet grounded), and any contradictions between user types. Keep it to what carries a decision.
These needs are the opportunities for /layers-product-strategy. If they're mostly assumed, consider /layers-observed-behaviour to gather evidence before building strategy on them.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

