Google Analytics logo

Google Analytics

Organization
runxhq
google-analytics

Read and normalize GA4 properties, metadata, standard reports, and realtime reports through any compatible Runx connector, preserving the measurement caveats an operator needs before turning analytics into action.

Overview

Publisherrunxhq
Repositoryrunx
Skill namegoogle-analytics
Stars
105
Forks
109
Bundled files
3
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 3 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by runxhq on GitHub. Read the source before you install it.

Installation

Install the Google Analytics AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/runxhq/runx/tree/main/skills/google-analytics
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/runxhq/runx.git /tmp/runx
mkdir -p .claude/skills
cp -r /tmp/runx/skills/google-analytics .claude/skills/google-analytics
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Google Analytics in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Google Analytics on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Google Analytics is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Google Analytics

Use this skill for bounded GA4 evidence: finding accessible properties, checking the dimension and metric catalog, measuring acquisition and on-site outcomes, or reading a short realtime window after a deployment or campaign.

The portable skill owns the measurement procedure and evidence contract. It does not own OAuth, credentials, tenants, connector selection, or provider routing. Operators may bring a local, self-hosted, third-party, or Runx-hosted connector. Any of those is compatible when it implements the operations and scopes below. Never place a connection id, credential handle, provider-config key, service URL, or tenant id in a skill input, fixture, or evidence packet; those bindings belong in the operator's Runx grant and connector configuration.

Read the right thing

Use properties when the property is unknown. Use metadata before composing an unfamiliar report: GA4 dimension and metric compatibility is part of the query contract, not a formatting concern. Use report for date-bounded analysis and realtime only for a current operational pulse.

normalize-report accepts a GA4-shaped result already supplied by a trusted import or connector test. It performs the same semantic checks, but its output says provider_status: not_called and source_status: supplied_result. A supplied packet is useful evidence with external provenance; it is never proof that Runx reached Google.

Use an explicit empty dimensions list for a metric-only total. Its response must have empty dimension headers and row dimension values; metric headers and values still match the exact ordered request.

A report should name the exact property, date ranges, ordered dimensions, ordered metrics, filters, ordering, currency, row limit, and offset needed for the decision. Keep it narrow. Wide reports are more likely to combine incompatible fields, hit quotas, hide detail in an (other) row, or give an agent an attractive but irrelevant aggregate.

The normalized packet maps each ordered value to its header name, retains row count and pagination, and carries timezone, currency, quota, and response metadata. Header drift is a hard failure because a row cannot be interpreted safely when the requested and returned fields differ.

Measurement caveats

Google Analytics is behavior and outcome evidence, not search-ranking evidence. The Organic Search channel can include search engines other than Google and depends on the property's channel definitions. Landing-page sessions and key events help quantify outcomes, but they do not explain why visibility changed. Join GA4 with Search Console on a deliberate page and date boundary, then inspect content, releases, attribution settings, and business context before making a causal claim.

Treat thresholding, sampling metadata, schema restrictions, and data_loss_from_other_row as material. The skill returns decision: usable_with_caveats when any is present. Do not silently compare a thresholded segment with an unthresholded total, add rows across overlapping date ranges, or treat a paginated page as the complete property.

Timezone, currency, reporting identity, key-event configuration, consent mode, and attribution settings can change interpretation without changing the query. Carry those assumptions into any comparison. Realtime data is provisional and suited to detection, not settled performance reporting.

Connector capability contract

A compatible connector exposes provider google-analytics with:

  • properties.list under properties.read, projecting properties.
  • metadata.get under metadata.read, projecting the property, dimensions, metrics, and fetch time.
  • reports.run under reports.read, projecting property, request, dimension and metric headers, rows, row count, pagination, response metadata, property quota, and fetch time.
  • reports.run_realtime under reports.realtime.read, using the same projected report shape and an explicit realtime request.

The connector decides how credentials are acquired and refreshed. It must enforce the caller's grant, bind the requested property, preserve Google errors, project only declared fields, and keep secret material out of results and receipts. This package is read-only; analytics administration, audience changes, event configuration, and deletion requests require separate governed capabilities.

Operator handoff

A sealed report packet is ready for review or for a higher-level organic-growth comparison. Preserve the evidence digest and the exact query when deriving trends. A useful organic acquisition slice commonly includes a landing-page dimension, a channel or source dimension, and only the outcome metrics needed for the decision, such as sessions, engaged sessions, key events, or revenue.

Stop when the property is ambiguous, requested fields are incompatible, response headers drift, pagination is incomplete for a whole-property claim, privacy metadata undermines the comparison, timezone or currency is unknown where it matters, the supplied result lacks provenance, or the configured connector lacks the exact capability.

The deterministic module exists for GA4 domain semantics that generic Runx tools cannot infer: binding ordered headers to values, validating report identity, parsing numeric metrics, and carrying privacy and completeness caveats. Runx owns provider execution, grants, digesting, projection, and receipts.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Google Analytics AI skill do?

Read and normalize GA4 properties, metadata, standard reports, and realtime reports through any compatible Runx connector, preserving the measurement caveats an operator needs before turning analytics into action.

Why use Google Analytics on TypingMind?

Because you install it once and use it with any model. Google Analytics is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Google Analytics in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/runxhq/runx/tree/main/skills/google-analytics. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Google Analytics?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Google Analytics?

As many as you like. As long as a model supports skills, you can use Google Analytics with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Google Analytics AI skill free?

Yes. It is published on GitHub by runxhq under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇