form-friction-finder
Skill2 starsChanged 4 months agoMITView on GitHub
Finds the fields, validation rules, error states and steps that stop people finishing a form they intended to finish. Use when form starts are healthy and submits are not, or when a field was added and completion dropped.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- mardab96/b2b-lead-generation-claude-skillstheir own version, first seen Jul 27, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- mardab96/b2b-lead-generation-claude-skills★ 5same name
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 repository26 other files it gives its agents.
- AGENTS.mdAGENTS.md
- above-the-fold-clarity-reviewSkill
- accessibility-conversion-blocker-checkSkill
- comparison-page-positioning-reviewSkill
- conversion-leak-finderSkill
- cta-clarity-checkSkill
- google-ads-landing-page-experience-reviewSkill
- hero-section-diagnosisSkill
- landing-page-ab-test-readoutSkill
- landing-page-copy-readability-passSkill
- landing-page-scale-readiness-checkSkill
- landing-page-triageSkill
- lead-form-sales-handoff-checkSkill
- mobile-conversion-reviewSkill
- objection-map-builderSkill
- offer-clarity-diagnosisSkill
- page-length-fit-checkSkill
- page-speed-impact-reviewSkill
- paid-traffic-message-match-auditSkill
- popup-and-overlay-timing-reviewSkill
- pricing-page-clarity-reviewSkill
- social-proof-strength-auditSkill
- thank-you-page-opportunity-auditSkill
- traffic-temperature-match-reviewSkill
- trial-vs-demo-path-decisionSkill
- trust-signal-auditSkill
The file
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 7 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
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
- Distinctive: line 22
If starts and submits are both unavailable, say the ranking is structural rather than measured.
- Distinctive: line 29
No device split on the form: test it on a phone anyway.
- Very specific: line 31
Not for: whether a field should exist at all (lead-form-sales-handoff-check), or the hesitation right before submitting …
- Distinctive: line 38
Check mobile input types, dropdown length, error messages, field order and visible privacy cues.
- Distinctive: line 49
Do not recommend a multi-step redesign as the first fix.
- Distinctive: line 72
Do not edit the form, validation rules or the CRM.
- Distinctive: line 73
Do not recommend a multi-step rebuild before the cheap fixes.
- Distinctive: line 74
Do not claim a completion-rate lift.
≈ “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.
Files
Form Friction Finder
Use this skill when
- the form's fields, validation behaviour and step structure are available, and starts exceed submits by more than the team can explain.
- a field, a validation rule or a step was added and completion dropped.
- people submit the form with obviously fake values in one particular field.
This skill asks why people stop filling a form in. Whether a field should exist at all belongs to lead-form-sales-handoff-check.
Required input
- the full field list with types, required flags, validation rules and order.
- the step structure: one page, multi-step, progressive, and what the progress indicator says.
- error behaviour: when errors appear, where, and what they say.
- numbers if they exist: form starts, submits, per-field drop-off, time to complete, device split.
- If starts and submits are both unavailable, say the ranking is structural rather than measured. The defects are still real; their order is not established.
If the data does not exist
- No form starts: GA4 fires
form_startandform_submitautomatically for many forms under enhanced measurement. Reports, then Engagement, then Events. If they are there, you have both numbers already and did not know it. - If they are not: the manual version is to fill the form yourself on a phone, wrongly on purpose. Submit with a blank required field, a locally formatted phone number, a plus-address email. Most of what this skill finds surfaces in that one pass, and it needs no analytics at all.
- No per-field drop-off: almost nobody has this. Say the ranking is structural and name the two candidates you would test first rather than pretending to an order you cannot see.
- No device split on the form: test it on a phone anyway. Form defects concentrate there.
Not for: whether a field should exist at all (lead-form-sales-handoff-check), or the hesitation right before submitting (trust-signal-audit).
Analysis workflow
- List fields, steps, required answers, validation, hidden fields, consent copy and submission behavior.
- Classify each field as essential, qualification, routing, compliance or nice-to-have.
- Compare friction level with traffic intent, offer value and buyer stage.
- Check mobile input types, dropdown length, error messages, field order and visible privacy cues.
- Recommend remove, reorder, make optional, split, prefill or explain fields.
Decision rules
- Everyone who started intended to finish. Treat the gap between starts and submits as mechanics, not motivation, and look for the mechanical cause.
- Validation that rejects a correctly typed real value is the highest-severity class in this skill. A phone rule that refuses a locally formatted number fails the most common way a person types their own number.
- Errors surfaced only on submit, plus a scroll away from the field, is the most abandonment-heavy pattern there is. Rank it above field count every time.
- A placeholder is not a label. It disappears when typing starts, which produces wrong entries as well as abandonment.
- Field count matters less than field cost. One field asking for something the visitor must go and look up costs more than three they can answer from memory.
- Per-field drop-off data outranks every heuristic here. When it exists, rank by it and say so; when it does not, say the ranking is structural.
- Do not recommend a multi-step redesign as the first fix. It is a rebuild, and the cheap fixes usually have not been tried.
Output format
| Field or behaviour | Defect class | Evidence | Severity | Fix |
| Field name or the rule | Validation / labelling / error handling / cost to answer / order | Quoted rule, number, or missing-data marker | High / medium / low | The smallest change |
End with:
Largest single defect:which one, and what it is estimated to cost, or "cannot rank without per-field data"Cheap fixes:the ones deployable this weekDecision:fix validation / fix error handling / add labels / reorder / ask for per-field dataMissing data:what would turn the structural ranking into a measured one
Practical example
User: "Form starts 610, submits 190. Fields: name, email, phone, company, website, how can we help. Phone validation rejects anything without a country code. Errors appear only after pressing submit and the page scrolls back to the top."
Assistant should note: 190 of 610 finishing means 420 people decided to fill this in and then stopped, which is a mechanics problem rather than an intent problem; validation that rejects a locally formatted phone number fails the most common way a person types their own number, and it fails silently until submit; errors surfacing only on submit, combined with a scroll to the top, means the user loses their place and has to hunt for what broke, which is the single most abandonment-heavy pattern in the list. Then: inline validation at field level and accepting local phone formats are the two fixes that address the largest share, and the field count question is secondary to them. The per-field drop-off data that would rank fields properly was not supplied.
Guardrails
- Do not edit the form, validation rules or the CRM.
- Do not recommend a multi-step rebuild before the cheap fixes.
- Do not claim a completion-rate lift.
- Say when the severity ranking is structural rather than measured.
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.

