agentsclimarketplace

Running growth experiments

Skill Giving-Back-Studio/growth-agent/skills/running-growth-experiments

Growth Agent — open-source skills that turn the AI you already use into a growth partner for your business, powered by Earth Care Network

Install
npx -y skills add Giving-Back-Studio/growth-agent --skill running-growth-experiments

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

3 things to look at

  • 12 days oldThe repository was created 12 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Reviews an Earth Care Network Growth Plan and turns its recommendations into armed, running automations. Use when someone wants a growth-plan checkup, asks what is confirmed versus open, wants to know what to work on next, or wants to automate their sales follow-up.

SKILL.md

2.9 KB, as published. Nobody here has run it

Running growth experiments

The Growth Plan is the living deliverable: story → offer → icp → signals → prospects → moves → automation_plan. This skill audits it and closes the gaps.

Audit

get_growth_plan returns the plan plus real metrics. Report plainly which modules are confirmed, which are drafted but unconfirmed, and which are empty. Do not describe an empty plan as a start — say it's empty.

The metrics block is the honest read on whether anything is working. Content created but never approved is not content shipped; say that when it's true.

Close gaps in order

Modules compound — an automation_plan is only as good as the icp under it. Work the earliest open one first.

  • update_growth_plan_module — draft or revise. Only some modules are draftable here, and the parameter differs: content (a string) for story/icp, payload for targets/channel_bets/challenges/agentic_solutions. offer, signals, prospects, moves, and automation_plan are confirm-only. There is no value parameter — sending one writes nothing.
  • confirm_growth_plan_module — after the owner approves it. Accepts all seven.

If icp is thin, stop and use refining-your-icp rather than drafting a weak one inline.

Arm the automations

Confirming automation_plan returns 2-3 recommendations computed from that business's own confirmed ICP, watchlists, and leads. These are computed, not authored — narrate what comes back, don't invent additions.

create_workflow_from_template instantiates one as a draft. Templates include daily-lead-digest, signal-inbox-digest, daily-gtm-followups, sales-inquiry-pipeline, sales-follow-up-3day.

Say clearly: it is a draft, it does not run until they publish it, and nothing sends without their approval. A person who believes automation is live when it isn't will miss real leads.

Don't arm a template the plan didn't recommend. If they ask for one anyway, that's their call — just say it isn't grounded in their data.

Choosing what's next

When someone asks "what should I do?", answer from their data, not best practice in general:

  1. No confirmed ICP → that, before anything else.
  2. ICP confirmed, no watchlist → arm monitoring; it compounds while they sleep.
  3. Signals arriving, none triaged → triage; unread signals are worth nothing.
  4. Leads with no follow-up → drafts; the pipeline leaks here.
  5. All of the above healthy → then automate the repeated motion.

Name the one next action, not five. A person with five priorities has none.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.