Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add malloydata/publisher --skill storefront-conventionsgit clone --depth 1 https://github.com/malloydata/publisherWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/malloydata/publisher/storefront-conventions)<a href="https://agentmods.dev/skills/malloydata/publisher/storefront-conventions"><img src="https://agentmods.dev/badge/skills/malloydata/publisher/storefront-conventions/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/malloydata/publisher/storefront-conventions"><img src="https://agentmods.dev/badge/skills/malloydata/publisher/storefront-conventions.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00041 | $0.00690 |
| Opus 5.5 | $0.00016 | $0.00276 |
| Sonnet 5.5 | $0.00008 | $0.00138 |
| Haiku 4.5 | $0.00004 | $0.00069 |
Grade A, and why
storefront-conventions scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Storefront conventions
Everything here is a property of this package's model, not of Malloy. The numbers quoted were measured against the bundled data on 2026-09-04 and are there to show the size of each effect, not to be reused as answers.
The grain is a line item
order_items has one row per product sold on an order. 25,356 rows, 11,000
orders.
So order_item_count (count()) counts lines and order_count
(count(order_id)) counts orders. A question about orders that reaches for the
row count is wrong by a factor of about 2.3 here.
total_sales includes cancelled and returned lines
This is the one that bites. total_sales is sale_price.sum() over every row,
with no status filter:
| Scope | Revenue |
|---|---|
Everything (total_sales as defined) |
2,098,177.97 |
Excluding Cancelled and Returned |
1,905,657.16 |
9.2% of what the model calls revenue is a line that was cancelled or sent back.
Nothing in the field name or its #(doc) says so.
Decide which one the question wants, and say which you used. If it wants realised revenue, filter it yourself:
run: order_items -> {
where: status != 'Cancelled' and status != 'Returned'
aggregate: total_sales
}
status takes five values: Complete, Shipped, Processing, Returned,
Cancelled. Every line on an order shares that order's status, so filtering
lines by status does not split an order across groups.
return_rate is a share of lines
return_rate is returned line items over all line items. It is not the share of
orders that had a return. If the question is about orders, aggregate
order_count with a Returned filter against total order_count instead.
region is where the customer lives
regions joins through customers.state, so region is the customer's home
region. It is not a shipping destination or a sales territory. A question about
where goods went cannot be answered from this model.
margin is gross, and per line
gross_margin is sale_price - products.cost on a single line, and
margin_rate is total_margin / total_sales. Cost of goods only: no
discounts, no shipping, no returns netted out, no operating cost. Call it gross
margin in the answer, never "profit".
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 76 lines · 41 tokens per session scan A 4199d637759d
storefront-conventions is a skill published in the GitHub repository malloydata/publisher (118 stars, last pushed today), licensed MIT. It adds 41 tokens to every session and 690 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-13.
Other skills, from other repositories
ckjia-shopping
A shopping search and price-comparison tool for Taobao, JD.com, Tmall, and Pinduoduo, with image-based product recognition. It requires the ckjia-shopping service and its access key to be enabled.
reimagine-it-extract
Emit the content signals reimagine-it reads from an HTML file — title, anchors, proper nouns, dates, numbers, emails, links, source hex colors, and the derived palette — as JSON without generating a redesign. Use when the user says /reimagine-it extract, "what does the engine see in this page", "extract the palette"…
reusable-visualization
Build ONE reusable chart visualization component that receives its data and its settings from the host application instead of fetching them, and declares the fields and config options the host exposes to viewers. Use this whenever a single chart component is reused across many different queries rather than built for…
developing-in-lightdash
Use when reading, creating, and editing Lightdash dashboards and charts as JSON, including dashboard layout and chart-type-specific configuration.
sdk-features
Map Lightdash SDK capabilities to their host-UI names and the app-code wiring each needs. Use when the user asks about a feature by name (Inspect data, drill-down, exports, shareable URLs), when offering newly available features after a template upgrade, or when wiring a host-facing capability into the app.
upgrade-preflight
Checks whether a self-hosted Lightdash upgrade is safe to run, and reads the tooling's answer without over-reading it. Use when upgrading a self-hosted instance, planning a maintenance window, answering "is this upgrade safe", or recovering a failed, hung, parked or lock-stuck migration — covers lightdash…