Pricing Promotions logo

Pricing Promotions

OrganizationPopular
anthropics
pricing-promotions

Price changes within the store's guardrails, rate recommendations from demand, occupancy, or sell-through pace, promotions with their scope, depth, and dates, the markdown option for stock or capacity that is not selling, and the margin preview before any of it is staged. Not needed for explaining why revenue moved (performance-insights) or writing promotional copy (marketing-campaigns).

Overview

Publisheranthropics
Repositorycommerce-agents
Skill namepricing-promotions
Stars
3K
Forks
574
Bundled files
Instructions only
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.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

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

Installation

Install the Pricing Promotions 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.
  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/anthropics/commerce-agents.git /tmp/commerce-agents
mkdir -p .claude/skills
cp -r /tmp/commerce-agents/merchant-agent/skills/pricing-promotions .claude/skills/pricing-promotions
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Pricing Promotions 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 Pricing Promotions 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 Pricing Promotions 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.

Pricing and promotions

Below, "listing" means whatever this operation prices: a product, a room type on given dates, a plan, or a ticket tier.

A price proposal is a few figures the tools returned, one staged change, and its preview. The operator decides from the preview.

Where each figure comes from

  • Read get_pricing_context for each listing first: current price, cost and margin_pct, the allowed range, the two caps, and demand_signal.
  • State the caps before a figure: max_price_delta_pct bounds a permanent move and max_promotion_discount_pct bounds a promotion's depth. For an ask past a cap, name the cap and propose a figure inside it, in the text or as a chip, and stage that figure once the operator picks it; do not describe the over-cap version as allowed, stage it to see whether it passes, or stage the capped version in its place unasked.
  • Quote margins; do not compute them. margin_pct comes from pricing context, and margin_before_pct, margin_after_pct, and margin_impact come from the staged change record, so a staging note carries no margin claim.
  • Write money in the listing's currency exactly as the tools returned it.
  • Standing repricing rules and price bands belong to the host's configuration and reach you as the range and caps pricing context returns; min_price_basis says whether the floor is the item's cost or a store rule, so say which when a floor decides the answer. Your own moves are staged one change at a time.
  • A listing with options is priced per variant: pricing context on the listing returns a row per variant, and each item you stage names a variant id.

The size of the move

  • Anchor on the operator's stated goal (clear a slow line, lift margin, hold volume, fill particular dates) and propose the smallest move that plausibly meets it.
  • State the expected effect as an expectation drawn from demand_signal or the pace figures, with the basis named.
  • Offer the options that are not a cut first when they fit: a shorter window, a narrower scope, or holding where it is.
  • For a slow mover handed over from inventory-operations, add the margin room from pricing context to the units, age, and pace it arrived with, say which way the numbers point, and stage the markdown only if the operator picks it.

Promotions and date-bound moves

  • Stage a promotion once it has a scope, a depth, and an end date; raise an open-ended or storewide discount as a question instead of filling it in.
  • A directed price move whose window has no dates yet ("for the spring push") is a price update: stage it now with the assumption in its note, and offer to convert it into a date-bound promotion once the dates are in hand.
  • Let the scope choose the tool: a move limited to particular dates or nights is stage_promotion, a lift included; stage_price_update moves the base price from now on, and is the tool only when that is what was asked.
  • Anchor a date-bound recommendation on occupancy or sell-through pace for the dates in question, never on a season average, and give it per date range or tier with the dates or days-out figure driving it. Leave the ranges the pace does not support alone.
  • Take weekday names from the returned dates: check them before writing "Sat-Sun" beside a window, and give the dates alone when unsure.
  • Point out in the reply any included item the promotion would sell under its floor.

Stage and preview

  • Every move goes through stage_price_update or stage_promotion; the preview shows before and after per item and the margin impact.
  • Make the staging note one sentence on why the change is safe or worth making; a present_change_preview headline and note say the same, one sentence each.
  • The guardrails you will meet here are the per-change item cap, the movement caps, and protected fields such as a regulated fee. Because promotions end, the alternative to a permanent move over max_price_delta_pct is a date-bound promotion; offer it for the operator to choose, and do not stage it in the move's place unasked.
  • Deliver a rate recommendation as a component: a staged change with its preview, or the figures in present_metrics with the drivers in a sentence or two before it.

Frequently asked questions

What does the Pricing Promotions AI skill do?

Price changes within the store's guardrails, rate recommendations from demand, occupancy, or sell-through pace, promotions with their scope, depth, and dates, the markdown option for stock or capacity that is not selling, and the margin preview before any of it is staged. Not needed for explaining why revenue moved (performance-insights) or writing promotional copy (marketing-campaigns).

Why use Pricing Promotions on TypingMind?

Because you install it once and use it with any model. Pricing Promotions 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 Pricing Promotions in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/anthropics/commerce-agents/tree/main/merchant-agent/skills/pricing-promotions. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Pricing Promotions?

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 Pricing Promotions?

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

Is the Pricing Promotions AI skill free?

Yes. It is published on GitHub by anthropics 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 👇