agentleFS
pooriaarabpooriaarab/skills/ads-lg/SKILL.md

ads-lg

Skill1 starsChanged 5 days agoMITView on GitHub

Evaluate LG Ads campaigns when you need a documented access route, partner-supplied measurement, reporting verification, or a decision about unsupported API and pixel assumptions.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

26 rules. About 0 are common, 21 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈21

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Treat LG Ads as a managed integration until LG supplies a current contract or partner guide.

    line 13
  2. Distinctive:

    Use ad-conversion-hub for the canonical event envelope, consent gate, identity rules, retry policy, and adapter contract.

    line 18
  3. Very specific:

    Do not add a public LG script or invent LG_ADS_PIXEL_ID or LG_ADS_TAG_ID as vendor-defined settings.

    line 68
  4. Distinctive:

    Keep the snippet behind the hub’s measurement consent gate.

    line 73
  5. Distinctive:

    Do not treat a browser request as proof that LG credited a conversion.

    line 74
  6. Distinctive:

    Keep the hub event taxonomy local until LG supplies a campaign-specific measurement contract.

    line 87
  7. Distinctive:

    After LG supplies written field names, map only the documented events and values at the adapter edge.

    line 101
  8. Very specific:

    Do not implement or probe a guessed route such as /conversions, /events, or /pixel.

    line 113
  9. Distinctive:

    Use this managed route instead:

    line 117
  10. Distinctive:

    Ask LG whether the campaign has a direct measurement integration or an approved measurement partner.

    line 119
  11. Distinctive:

    Keep dispatch disabled until the documents identify the exact request.

    line 128
  12. Distinctive:

    Use the hub’s consent gate before measurement or identity processing:

    line 145
  13. Distinctive:

    Do not hash email or phone for LG merely because another ad platform uses SHA-256 matching.

    line 151
  14. Distinctive:

    Do not upload customer lists or create lookalike audiences without a documented LG product, permitted field, and consent basis.

    line 153
  15. Very specific:

    Do not create lg_click_id, lgclid, or another guessed parameter.

    line 166
  16. Distinctive:

    Ask the LG team or measurement partner whether the approved campaign URL may carry a campaign code or UTM values.

    line 168
  17. Distinctive:

    Never require a click ID before recording a confirmed signup, purchase, subscription, or refund in first-party systems.

    line 173
  18. Distinctive:

    Treat these statements as product capabilities, not as an integration schema.

    line 188
  19. Distinctive:

    Start with one approved market, one audience, one creative, and one outcome definition.

    line 195
  20. Distinctive:

    Put the campaign ID, approved measurement method, consent basis, and report owner in the campaign record.

    line 197
  21. Very specific:

    Keep the LG adapter in skipped state until the contract and credential are present.

    line 201
  22. Distinctive:

    Do not guess budget floors, bid fields, learning thresholds, placements, currency rules, or reporting delays.

    line 204
  23. Distinctive:

    Platform proof: use the provisioned LG Insights Dashboard or the partner’s documented campaign report.

    line 213
  24. Very specific:

    If LG provides no report, test-event view, or documented response, return skipped or failed according to the hub adapter contract.

    line 225
  25. Distinctive:

    Keep credentials in the server secret store.

    line 241
  26. Distinctive:

    Keep canonical events, consent, identity normalization, click storage, retry policy, and payment reconciliation in the hub.

    line 249

≈ “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.

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

LG Ads

LG Ad Solutions is a connected-TV and cross-screen advertising offering. Its public site describes LG inventory, audience data, media, and outcome measurement, then directs advertisers to its team. See LG Ads and Solutions.

Treat LG Ads as a managed integration until LG supplies a current contract or partner guide. The public materials checked on 2026-08-30 do not publish a self-serve advertiser signup, a public ads API, or a public conversions API. The real entry point is Contact LG Ads.

Use ad-conversion-hub for the canonical event envelope, consent gate, identity rules, retry policy, and adapter contract. This skill records only LG-specific facts that the vendor publishes. Pair it with ad-experiments for one-audience tests, seed sizing, PII-export authorization, and payment-provider truth.

Account and access

The official advertiser route is a contact request. LG says that its team can help advertisers reach connected-TV audiences and measure cross-screen campaign outcomes. See Contact LG Ads.

There is no public self-serve advertiser signup in the official pages checked for this skill. The public Contact page is the documented access route. Do not create a guessed advertiser account, developer account, token, or API client. Request access through LG’s team instead.

Include these questions in the request:

  1. Which countries, inventory types, and buying routes are available for the campaign?
  2. Which team or partner provisions campaign access and reporting access?
  3. Does the campaign include a website beacon, app integration, or another measurement method?
  4. Which event fields, identity fields, retention periods, and deduplication rules apply to this campaign?
  5. Which report or dashboard proves receipt, attribution, and final results?

LG’s public site lists one-stop planning, activation, and measurement across viewing platforms. That does not establish a public API or a self-serve buying flow. See LG Ads and LG’s platform announcement.

LG also operates an Insights Dashboard. Its public description advertises integrated analytics, live digital campaign performance, and closed-loop attribution, but the page presents sign-in rather than public advertiser registration. Use it only after LG provisions access.

Client-side web beacon

LG Ads does not publish a public browser-tag snippet, pixel ID, tag ID, or installation contract in the official materials checked for this skill. See LG Ads and Solutions.

The related Alphonso privacy policy says that some media partners may place an Alphonso web beacon, also called a pixel, on advertisements or websites. It says that these beacons help partners measure ad effectiveness and website visitation. The same policy directs advertisers to visit Alphonso’s site or email Alphonso for service access.

Therefore:

  • Do not add a public LG script or invent LG_ADS_PIXEL_ID or LG_ADS_TAG_ID as vendor-defined settings.
  • Install a beacon only when LG or its measurement partner supplies the current snippet, source origin, identifier, allowed pages, and consent terms in writing.
  • Keep the snippet behind the hub’s measurement consent gate.
  • Do not treat a browser request as proof that LG credited a conversion.

The privacy policy says that an Alphonso beacon uses JavaScript. Load only the exact code supplied through the approved partner route. Do not copy a snippet from a blog, an old case study, or browser network traffic.

Rule setup and event mapping

LG’s public pages describe outcome measurement, including web traffic and purchase measurement, but they do not publish LG event names, rule IDs, JSON fields, or a conversion setup API. See LG’s ACR measurement overview.

Keep the hub event taxonomy local until LG supplies a campaign-specific measurement contract. Do not send these names as LG event names:

Hub eventLG Ads action before partner documentation
page_viewRecord in first-party analytics only.
view_contentRecord in first-party analytics only.
leadRecord the qualified lead in the source system.
signupRecord the account creation in the source system.
begin_checkoutRecord the checkout start in the source system.
purchaseReconcile the payment-provider charge in the source system.
subscription_startReconcile the active subscription in the source system.
refundReconcile the payment-provider refund in the source system.

After LG supplies written field names, map only the documented events and values at the adapter edge. Keep the canonical event names and consent rules in ad-conversion-hub.

Server-side conversions API

LG Ads has no public conversions API contract in the official material checked for this skill. No public LG endpoint, authentication header, token scope, payload schema, event-name set, hashing rule, timestamp unit, retention window, or deduplication field was published in those materials. See LG Ads, Solutions, and Contact LG Ads.

Do not implement or probe a guessed route such as /conversions, /events, or /pixel. Do not use https://lgads.tv/ as a token-validation endpoint. A web page response cannot prove that a conversion was accepted.

Use this managed route instead:

  1. Ask LG whether the campaign has a direct measurement integration or an approved measurement partner.
  2. Request the current endpoint, authentication method, required fields, allowed identifiers, response contract, retry guidance, and deduplication behavior in writing.
  3. Confirm whether the integration supports browser events, server events, or reporting only.
  4. Store any resulting credential in the server secret store under a project-chosen name. Do not represent that name as an official LG field.
  5. Keep dispatch disabled until the documents identify the exact request.

LG’s own measurement material says that advertisers should contact LG to discuss campaign goals and return-on-ad-spend measurement. See ACR measurement.

Identity and consent

The Alphonso privacy policy says that its service collects device data only when a viewer agrees to participate. It also says that viewers can withdraw that consent in the settings menu of the television where the service is integrated. See Smart TV privacy.

That policy says that audio signatures are associated with an IP address or device ID, but not with a name, postal address, or email address. Do not infer that LG Ads accepts email, phone, IP, device, or household identifiers for a campaign without LG’s current written terms. See Smart TV privacy.

Use the hub’s consent gate before measurement or identity processing:

  • Require measurement: true before loading a consented client measurement asset or dispatching an event.
  • Require ad_user_data: true before processing identifiers for an approved LG integration.
  • Do not hash email or phone for LG merely because another ad platform uses SHA-256 matching.
  • Do not upload customer lists or create lookalike audiences without a documented LG product, permitted field, and consent basis.

LG’s technology page describes ACR-based viewership data, content and service targeting, purchase-habit segments, ad-exposure segments, and geographic targeting. These are audience-product descriptions, not proof of a customer identity API. See LG technology.

Click IDs and first-party storage

LG Ads does not publish an LG-owned click parameter, cookie name, attribution window, or click-ID retention period in the official materials checked for this skill. See LG Ads and Contact. Do not create lg_click_id, lgclid, or another guessed parameter.

Ask the LG team or measurement partner whether the approved campaign URL may carry a campaign code or UTM values. If approved, capture the value on your landing request and store it under the hub’s first-party click rules. Keep first-touch and most-recent values only when the measurement design needs both.

Never require a click ID before recording a confirmed signup, purchase, subscription, or refund in first-party systems. Do not claim that LG attributed the event unless the LG report or partner response proves it.

Audience and measurement boundaries

LG describes ACR as first-party technology that captures viewing behavior on LG Smart TVs across linear television and OTT. Its technology page lists segments based on content, devices and services, purchase habits, ad exposure, viewing habits, and location. See LG technology.

LG also describes cross-device placements and measurement KPIs such as incremental reach, frequency, share of voice, attribution, and brand lift. See LG solutions.

Treat these statements as product capabilities, not as an integration schema. The cited pages do not supply a public upload endpoint, audience-file format, click-ID join key, or conversion postback contract. See LG technology and Contact.

Managed launch checklist

  • Start with one approved market, one audience, one creative, and one outcome definition.
  • Put the campaign ID, approved measurement method, consent basis, and report owner in the campaign record.
  • Confirm whether a partner or LG team owns the beacon, event feed, or report.
  • Define first-party success using payment, signup, or subscription truth.
  • Keep the LG adapter in skipped state until the contract and credential are present.

Do not guess budget floors, bid fields, learning thresholds, placements, currency rules, or reporting delays. Ask LG for the campaign-specific values.

Verification

Use three proofs:

  1. Request proof: record the approved onboarding record and any documented partner response. Redact credentials and identifiers.
  2. Platform proof: use the provisioned LG Insights Dashboard or the partner’s documented campaign report. Confirm campaign identity, dates, delivery, and the agreed outcome metrics. LG describes its reporting as live campaign performance and closed-loop attribution in the Insights Dashboard.
  3. Business proof: reconcile reported outcomes with first-party signup, payment, subscription, and refund records. The hub remains the source of business truth. See ad-conversion-hub.

A successful page load, beacon request, dashboard login, or partner HTTP response does not by itself prove ad attribution. Require the platform report or documented partner result for that claim.

If LG provides no report, test-event view, or documented response, return skipped or failed according to the hub adapter contract. Do not label an event as sent merely because first-party analytics recorded it.

Common pitfalls and security

  • Treating LG’s public marketing pages as API documentation.
  • Treating the Insights Dashboard as a public API.
  • Inventing a pixel ID, click ID, endpoint, token name, event field, or hash rule.
  • Sending raw or hashed customer identifiers before LG documents the field and the hub grants consent.
  • Treating ACR or television opt-in as consent for every website identifier.
  • Treating a beacon request as conversion proof.
  • Letting a missing LG partner integration block a successful payment webhook.

Keep credentials in the server secret store. Never place them in browser bundles, URLs, logs, screenshots, or commits. Load vendor code only from the official HTTPS origin or an approved partner asset. Apply the hub’s consent, retry, dead-letter, and failure-isolation rules.

Hub conventions

Pair this adapter with ad-conversion-hub. Keep canonical events, consent, identity normalization, click storage, retry policy, and payment reconciliation in the hub. Keep only the documented LG syntax in this adapter.

A missing partner contract or missing approved credential must produce a logged skipped result — the adapter has no route to dispatch. A malformed or missing required field on an otherwise-configured, enabled integration must produce a logged failed result instead, per the hub adapter contract, so a misconfigured but enabled integration cannot silently drop conversions. Neither result may fail the checkout or payment webhook.

Official sources checked (2026-08-30)

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.