Search the docs, videos, blog, apps, teams, skills and release notes.
At a glance
What it is
Get your products recommended by AI shopping assistants and answer engines. Test where you appear today, make your catalog readable by machines and earn mentions elsewhere.
A skill is a written procedure a bot loads when a job calls for it. This one is a single file, SKILL.md, and the whole file is on this page.
Each of these library teams gives its bots this skill. Import one and look inside first.
The whole file, exactly as it downloads. Read it before you rely on it.
SKILL.md
70 lines · 827 words
As of: 2026-05-16
Mode skill. Default-enabled on the Storefront specialist.
Use when the user wants to be found by AI shopping agents and generative answer engines, ChatGPT shopping, Perplexity, Gemini, and agentic-checkout flows that became material traffic sources in late 2025 / early 2026. Use alongside seo-organic (Beacon); this is the storefront-side complement.
Trigger phrases:
Get your products recommended by AI shopping assistants and answer engines. Test where you appear today, make your catalog readable by machines and earn mentions elsewhere.
Nothing. It ships with Murage, which is open source. Your bots run on the AI plan you already have.
Yes. Open it in Settings → Skills to read, edit or duplicate it. Skill Guard scans it again after every edit.
Download the app, connect the AI you already pay for, and tell your Chief of Staff what needs doing. Plan on about ten minutes from install to a working team.
No account needed · Runs on the AI plan you already pay for
Paste it into your agent's instructions (CLAUDE.md, AGENTS.md or a custom GPT), or add it to a bot in Murage.
Ask Ember for the bot, then switch the skill on in its settings.
Step 1
Send your Chief of Staff this message. It proposes the bot and its role first, and creates it when you say go.
Send this to Ember
“Create a bot that uses the Agentic GEO skill, and show me the plan first.”
You're done when your Chief proposes the bot and its role, and the bot appears in your sidebar once you say go.
Step 2
You're done when you see the bot’s skills and the library.
Step 3
You're done when the skill shows as on for that bot.
Step 4
You're done when the bot takes the job and opens the skill when it needs it.
Prefer to do it yourself? Choose + in the sidebar, then New Bot.Read the full guide to skills and Skill Guard
What it teaches a bot, what it covers, and the whole file to read.
The principle
**Mode skill.** Default-enabled on the Storefront specialist.
What it covers
When to use
Procedure
Decision rules
Anti-patterns
Before / after
"How do we show up in ChatGPT shopping results?"
"Optimize our products for AI search."
"We need to be discoverable to AI agents."
"Buyers ask AI for recommendations, how do we land in those answers?"
1. Establish the agentic baseline. Probe the surfaces first. Run 15-30 buyer-language queries through ChatGPT, Perplexity, and Gemini shopping. Per query, log: did the brand appear? Position? Framing? Against which competitors? This is rank-tracking for generative surfaces.
2. Make the catalog machine-readable. Agents read structured data more reliably than rendered HTML. Per priority SKU:
Product schema (JSON-LD): name, description, brand, sku, gtin, offers (price, currency, availability), aggregateRating, review.
Plain-prose description, declarative, specific, free of marketing fluff. Lead with what the product is.
Explicit "best-for" framing: who it serves, what use case, what it is not for. Agents extract these as filters.
Honest named comparisons ("comparable to X in feature Y; differs in Z"). Agents reason comparatively.
3. Optimize the product feed. Agentic checkout flows (Shopify's March 2026 agentic-storefront release; Amazon Rufus; Walmart Sparky) route via feeds, Google Merchant Center, Shopify's agent feed, Meta's shop feed. Audit feed fields: title (front-load brand + product type + key spec), description (200-500 words, plain language), price, availability, GTIN, high-res image URL, category, attribute fields (color, size, material, age group). Missing fields silently exclude you.
4. Earn citation surface outside your domain. Generative engines weight third-party signals heavily. Priority sources: published reviews on category authority sites, Reddit and category-forum mentions, comparison articles, YouTube reviews with transcripts, podcast mentions with show notes. A Reddit thread where a buyer asks "what's the best X for Y" and your brand is named with a clear reason often outweighs any on-site move.
5. Measure visibility, not rank. Generative surfaces lack stable position. Per query, track appearance rate across N runs, framing quality (positive / neutral / negative), citation source. Re-run monthly. Report trend, not point estimates.
Plain prose beats marketing copy on agentic surfaces. "Wool mid-layer designed for sub-zero hiking" beats "Conquer the elements with our premium wool mid-layer."
Structured data is the first move, always. No JSON-LD product schema = invisible to most agents. Fix this before any other GEO work.
Reddit and category forums are part of your SEO surface now. Engineering authentic presence in buyer communities is the highest-payoff GEO move for most brands.
As of 2026-05-16: Shopify's agentic-storefront layer (released March 2026) exposes a dedicated agent feed separate from the standard product feed. Brands not opted in are invisible to Shopify-mediated agent traffic. Check opt-in status before any other GEO work on Shopify stores.
Do not chase every agent. Prioritize the surfaces where your buyers actually research. If your category has zero presence on Perplexity but heavy presence on ChatGPT shopping, allocate accordingly.
Stuffing the product description with keywords the way 2015-era SEO did. Generative engines penalize that pattern and your description reads badly to humans too.
Treating GEO as a one-time project. Surfaces change monthly; this is operating cadence, not a launch.
Optimizing for AI surfaces while letting the human-readable PDP rot. The buyer still has to convert after the agent surfaces you.
Buying "AI SEO" services that promise rank guarantees on generative surfaces. There is no stable rank to guarantee.
Quoting any tactic from before mid-2025 without dating it as potentially stale. The surfaces did not exist in their current form then.
Ignoring Reddit and category forums because they "feel off-brand." That is where the agent's training data lives.
Brief: "We want to show up when people ask ChatGPT for products like ours."
Before (vague):
Add AI keywords to descriptions and write a blog post.
After (probe-first, structured, 2026-05-16):
Step one: run 20 buyer queries through ChatGPT shopping, Perplexity, and Gemini; log appearance rate as baseline. Step two: audit Product JSON-LD on top 10 SKUs, confirm name, description, gtin, offers, aggregateRating populate. Rewrite descriptions in declarative prose with "best-for" framing. Step three: confirm Shopify agentic-storefront opt-in (released March 2026; many stores default off). Step four: identify three forums and one subreddit where buyers ask comparison questions; build a 90-day authentic-presence plan with Copy. Measure: re-run the probe set at days 30, 60, 90; report appearance-rate and framing-quality deltas per query.
SKILL.md
70 lines · 827 words
The text of SKILL.md is shown as it downloads, under the Apache-2.0 license, by Ferrox Labs. The license is the one in the file's own frontmatter.
Your new bot
Agentic GEO bot
Waiting in your sidebar once you say go.
No account needed · Runs on the AI plan you already pay for