build-funnel
Skill1 starsChanged 3 months agoMITView on GitHub
Build or overhaul a conversion destination — landing page, sales page, marketplace listing, lead-capture form, or booking flow — instrumented and friction-audited. Run when strategy names a destination that doesn't exist or underperforms.
In this repository22 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- close-dealSkill
- connect-toolsSkill
- create-contentSkill
- define-positioningSkill
- design-offerSkill
- experimentSkill
- grow-referralsSkill
- grow-seoSkill
- health-checkSkill
- launch-campaignSkill
- launch-productSkill
- manage-pipelineSkill
- measureSkill
- plan-strategySkill
- research-marketSkill
- review-monthlySkill
- run-adsSkill
- run-outboundSkill
- run-partnershipsSkill
- setup-teamSkill
- tune-revopsSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
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 20
Test end-to-end: Every link, every form, a test transaction where possible.
- Very specific: line 31
No build platform reachable (no site, no budget, tools unauthorized) → use the step 3 fallback ladder.
- Distinctive: line 32
Test transaction fails or links break → fix before launch, no exceptions.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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
/build-funnel — Build Where the Traffic Lands
Inputs: knowledge/strategy.md (which destination, for which channel), knowledge/offer.md, knowledge/positioning.md, knowledge/research/customer-language.md, knowledge/tool-registry.md.
Agents: funnel-builder leads; copywriter writes; tools-operator executes in the platform.
Method: knowledge/methods/funnel-and-measurement.md, knowledge/methods/offer-design.md (presentation).
Steps
- Define the one action. What must a visitor do here (buy / sign up / book / message)? Everything else on the page is subordinate or removed.
- Confirm message match. What promise brings visitors here (from the campaign/channel)? The headline continues that promise exactly.
- Choose the build platform. Existing stack first; else tools-operator recommends per
/connect-tools. Marketplace listings build in the marketplace's own editor. Zero-budget / no-website fallback (never stall here): a free-tier page builder (Carrd, Canva site via the Canva MCP, Notion public page via the Notion MCP), a simple static HTML page the team writes and the operator hosts free (GitHub Pages/Netlify, guided steps provided), or — for conversation-led B2B — a booking link plus a one-page PDF instead of a website. Something live and testable this week beats something perfect next month. - Draft to the outline.
templates/landing-page-outline.md, copy by copywriter from the customer-language corpus, offer and claims verbatim from their sources. Pull fresh proof (testimonials, reviews, results) fromknowledge/research/social-proof.md— real proof on the page outperforms any adjective. - Friction audit. Steps counted and challenged, forms minimized, mobile checked, price visible before final step, load speed sane.
- Instrument. Conversion event defined, analytics connected, UTM conventions applied to inbound links.
- Test end-to-end. Every link, every form, a test transaction where possible. Broken funnels are worse than no funnel — they spend trust.
- Approval & launch. Public changes fall under the publishing gate; new spend under the money gate. Then live, first-hour check, baseline recorded.
Before you start
- Required inputs:
knowledge/strategy.mdapproved (active),knowledge/offer.md+knowledge/positioning.mdatseeded+ — a funnel without an approved offer converts to nothing. - Scan
knowledge/decision-log.mdfor tagschannel,offer,positioning— the page states only approved claims. - Gate check: public changes = publishing gate; new platform spend = money gate. Claims Review Checklist (
standards/compliance.md) runs before the publishing approval.
If things go wrong
- No build platform reachable (no site, no budget, tools unauthorized) → use the step 3 fallback ladder; if even that's blocked, park exact guided steps in
knowledge/pending.md. - Test transaction fails or links break → fix before launch, no exceptions; a broken funnel spends trust.
- Instrumentation impossible on the platform (e.g., marketplace) → use the platform's native stats and record the measurement limitation in the baseline note.
Outputs
A live, tested, instrumented destination; its URL/location and baseline noted. Close with the report per templates/skill-report.md.
Writeback
Destination + config to knowledge/tool-registry.md; build decisions (tag: channel) to knowledge/decision-log.md; baseline to knowledge/results-history.md.
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.

