agentleFS
BagelHoleBagelHole/BagelSkills/skills/ux-friction-hunter/SKILL.md

ux-friction-hunter

Skill2 starsChanged 4 months agoMITView on GitHub

Use this skill when evaluating a live or locally runnable product through real user journeys. Operate it like a skeptical first-time user, collect reproducible evidence for confusion and failure, distinguish defects from preferences, and rank UX friction by blocked outcomes rather than visual taste.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 8 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Test accounts, seed data, and forbidden side effects

    line 39
  2. Distinctive:

    If a required input is retrievable from available tools or artifacts, retrieve it rather than asking the user.

    line 42
  3. Distinctive:

    Write three to seven outcome-oriented missions without prescribing clicks.

    line 48
  4. Distinctive:

    Use the product based on visible cues before consulting internal knowledge.

    line 60
  5. Distinctive:

    Test loading, empty, validation, error, success, back navigation, refresh, duplicate action, narrow viewport, keyboard, and interruption …

    line 66
  6. Distinctive:

    Complete when: every observation has the correct claim strength

    line 74
  7. Distinctive:

    Do not perform destructive, financial, publishing, or external messaging actions unless explicitly authorized.

    line 84
  8. Distinctive:

    Do not infer analytics or user frequency from a single exploratory run.

    line 87

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

UX Friction Hunter

Outcome

Find the moments where real users hesitate, misunderstand, abandon, or fail—and make every finding reproducible.

This skill changes the agent's process: it requires evidence, explicit claim strength, and a checkable completion boundary. Load references/playbook.md only when deeper prompts or rubrics are needed. Use assets/output-template.md when producing the final artifact.

When to use

  • The user requests UX testing, product QA, conversion review, usability review, or dogfooding.
  • A critical journey such as onboarding, checkout, booking, publishing, or account recovery needs evaluation.
  • The team suspects users are dropping off but does not know where.

Do not use when

  • The user only requests visual restyling without journey evaluation.
  • The target cannot be run or observed and no artifacts are available.

Required inputs

  • Target URL/build and allowed test scope
  • Primary personas and jobs to be done
  • Test accounts, seed data, and forbidden side effects
  • Known device/browser requirements

If a required input is retrievable from available tools or artifacts, retrieve it rather than asking the user. Ask only when a missing answer represents authority, preference, or risk that tools cannot resolve.

Workflow

1. Define missions

Write three to seven outcome-oriented missions without prescribing clicks. Include the new-user path and at least one recovery or edge journey.

Complete when: missions and safety boundaries are explicit

2. Establish evidence capture

Record environment and starting state. Capture screenshots, exact copy, URLs, console/network evidence, and reproducible steps while testing.

Complete when: evidence can reproduce observations

3. Attempt journeys naively

Use the product based on visible cues before consulting internal knowledge. Record expectation at each decision point and the observed result.

Complete when: each mission has a recorded outcome and expectation gap

4. Probe state transitions

Test loading, empty, validation, error, success, back navigation, refresh, duplicate action, narrow viewport, keyboard, and interruption where relevant.

Complete when: applicable non-happy states have been exercised

5. Separate findings

Classify each observation as blocker, defect, friction, accessibility issue, content problem, or preference. A preference without an affected outcome is not a defect.

Complete when: every observation has the correct claim strength

6. Rank and report

Rank by user impact, frequency likelihood, confidence, and recovery cost. Provide concise reproduction, evidence, expected behavior, and the smallest credible remedy.

Complete when: priority findings have evidence and retest criteria

Operating rules

  • Do not perform destructive, financial, publishing, or external messaging actions unless explicitly authorized.
  • One finding should describe one causal problem, not a bag of observations.
  • Quote interface text exactly when wording causes the problem.
  • Do not infer analytics or user frequency from a single exploratory run.
  • Retest critical findings from a clean state when possible.

Output contract

The final result must contain:

  • Mission completion matrix
  • Ranked findings with severity and confidence
  • Reproduction steps and attached evidence
  • Expectation versus observation
  • Suggested remedy and retest criteria
  • Coverage and untested areas

Distinguish observed, inferred, assumed, and unverified claims. Link to files, commits, screenshots, commands, logs, or supplied sources when those artifacts support a consequential conclusion. Keep the executive result concise; move detail behind appendices when necessary.

Common pitfalls

  • Aesthetic review disguised as usability testing.
  • Happy-path tourism that never probes recovery states.
  • Vague findings such as “confusing UX” without a blocked outcome.
  • Severity inflation unsupported by likely user impact.

Verification checklist

  • The requested outcome—not merely the procedure—has been satisfied.
  • Every consequential claim is supported, qualified, or explicitly unverified.
  • The workflow's completion criteria are met in order.
  • Safety boundaries and forbidden side effects were respected.
  • The output follows assets/output-template.md or contains equivalent information.
  • Remaining unknowns have an owner, next action, or stated consequence.

Resources

  • references/playbook.md — deeper domain prompts and evaluation guidance.
  • assets/output-template.md — reusable final-output structure.
  • evals/evals.json — output-quality cases.
  • evals/trigger-evals.json — positive and negative description-trigger cases.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.