Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Retailgrid Pricing Agent

    1

    Run retail pricing from one prompt: build Grids, optimize margin, explain prices and approve changes in Retailgrid via MCP.

    Free

    1 useSecurity scanned
    Retailgrid Pricing Agent

    Retailgrid Pricing Agent

    Example session with this skill installed

    Maximize margin for the grocery category. Keep price index below 100 for KVI products.

    • Read your context and instructions
    • Compiled the retailgrid pricing agent

    Summary

    Built a Grocery Grid (1,842 products, confirmed by you), reused last week's product roles and ran a profit optimization with a KVI price-index guardrail. Example output, figures are illustrative.

    Grid: Grocery margin (1,842 rows, goal profit, horizon 30 days)

    Rules applied

    1. Margin at least 15%
    2. Optimizer price for profit
    3. KVIs at or below the lowest competitor price (price index 100 or less)
    4. No more than 8% change per step
    5. Price endings .99 or .49
    MetricForecast, 30 days
    Revenue-0.6%
    Gross profit+4.1%
    Units-1.2%

    Changes: 611 up, 148 down, 1,083 unchanged

    SKUProductCurrentProposedChangeReason
    10234Olive oil 1 L8.498.99+5.9%Low elasticity, competitors at 9.20
    31877Basmati rice 2 kg4.494.79+6.7%Margin below category, stable demand
    20871Milk 1 L (KVI)1.291.19-7.8%KVI rule, lowest competitor 1.19
    20945Bread 500 g (KVI)1.491.39-6.7%KVI rule, price index was 107

    Flags: 37 rows have no unit cost and were left unchanged.

    Next steps

    • Approve the 611 increases, or only the top 100 by gross profit
    • Review the 37 rows with missing cost
    • Say "publish" when you want accepted prices sent downstream

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Retailgrid is an AI pricing workspace for mid-market retailers. This skill connects your agent to the Retailgrid MCP server and turns one plain sentence into a full pricing run.

    "Maximize margin for grocery. Keep price index below 100 for KVI products."

    Your agent finds or builds the right Grid, sets up pricing rules, runs price optimization, explains each recommendation and walks you through approval. Nothing is created, approved or published without your explicit yes.

    What it does

    • Builds Grids from catalog filters: category, brand, store, channel or sales metrics, with a row-count preview first.
    • Translates constraints into rules: margin floors, competitor price index, max change per step, price endings, size and family relations.
    • Runs optimization for profit or revenue and reports impact on revenue, gross profit and units over 7 to 180 days.
    • Ranks the biggest moves with brand caps and up/down quotas.
    • Explains any price: which rules applied, the binding constraint, the demand forecast, plus what-if prices that save nothing.
    • Classifies product roles such as KVIs and traffic builders.
    • Approves, rejects, overrides and publishes, only when you ask.

    Built for safe use

    • Every write is confirmed with you first, with the exact Grid, rows and rules shown.
    • Publishing is a separate step and only runs when you say "publish".
    • Idempotent request IDs prevent duplicate actions on retries.
    • Fixed report format: rules applied, impact, top moves and what needs your decision.

    What you need

    • A Retailgrid account with AI connections enabled. Ask at retailgrid.io/contact, or book a demo at retailgrid.io/demo.
    • The Retailgrid MCP connector: https://mcp.retailgrid.io/mcp (OAuth sign-in, no API keys). Setup steps for Claude, Claude Code, ChatGPT and other MCP clients are included.

    What's inside

    • SKILL.md: the workflow and ground rules
    • references/setup.md: connecting the MCP server
    • references/tools.md: all 41 tools, read vs write
    • references/rules.md: pricing language to rule mapping
    • examples/: a full worked run

    Only works with a Retailgrid account where AI connections are enabled for the organization. The agent sees only Grids and jobs owned by the signed-in user. Live formula columns cannot be read or sorted through MCP.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    • Free with an Agensi account

    Requires

    Retailgrid
    Retailgrid account

    Permissions

    Network Access

    Allowed Hosts

    retailgrid.io

    All data access goes through the Retailgrid MCP server, which the user connects and authorizes with OAuth. The skill runs no scripts, reads no local files or environment variables, and contains no credentials.

    Listed1 month ago

    What's inside

    Frequently Asked Questions

    Yes. The skill works on your own Retailgrid data, and AI connections must be enabled for your organization. Ask at retailgrid.io/contact, or book a demo at retailgrid.io/demo.

    Add https://mcp.retailgrid.io/mcp as a remote MCP server in your agent, or pick Retailgrid in the Claude connectors directory. You sign in with your Retailgrid login in the browser. No API key is needed. Steps for Claude, Claude Code, ChatGPT and other clients are in references/setup.md.

    No. Every write is confirmed with you first. Approving a proposal never publishes it, and publishing runs only when you say "publish". Read and write access are also separate OAuth scopes, so you can connect read-only.

    Margin and markup floors, competitor price matching and price index targets, max change per step, fixed prices, price endings, size and family price relations, and price sync across a group. Rules can target categories, roles such as KVIs, stores, zones or channels.

    Any agent that supports SKILL.md and remote MCP servers with OAuth, including Claude, Claude Code, ChatGPT, Cursor and Codex.

    No. Data flows only between your agent and the Retailgrid MCP server you authorize. The agent sees only Grids and jobs you own, every call is checked against your organization's permissions, and you can disconnect any time in Retailgrid under Settings > AI connections.

    New in E-commerce