MCP App Store
by KeenEthics
Laurence icon

Laurence MCP Server for ChatGPT

By Laurence

Overview

Laurence runs Amazon advertising for e-commerce brands: models set the bids every hour and people review the work. This connector gives your AI assistant read-only access to your company's Laurence data, so you can ask about your Amazon business in your own words. What you can ask about: Sales: total and daily revenue, units, net profit, product margins and customer acquisition cost. Advertising: spend, ad sales and returns by campaign, product, placement and hour; branded and non-branded ads; campaigns, ad groups, keywords and negative keywords as they stand on Amazon. Search: the search terms shoppers used, market search volume and share, search term popularity and keyword clusters. Products: catalog details, Best Sellers Rank history, FBA inventory, and deals and coupons. What Laurence did: keywords added, bid and budget changes, the full history of one target, and the account's shared Slack channel with Laurence. Who it is for: brands with a Laurence account. You need a Laurence login with access to at least one company. On first use you sign in to Laurence and choose the one company the assistant may read. Read-only: every tool only reads. Nothing can change bids, budgets, campaigns or listings, and nothing is sent to Amazon. Limits: answers cover the stores of the company you chose, and are as current as Laurence's latest data.

How to connect

  1. ChatGPT: Open the Laurence listing in the ChatGPT app directory and choose Connect. ChatGPT listing Laurence exposes 47 tools here.

Tools

explain_page

ChatGPT
Explain one page of the Laurence platform: its controls and sections in the order they appear, and for each number where it sits, what it means with its formula, and which data tool reproduces it. It answers what a page or a number on it means and how a page works. It never returns the account's numbers; the explanation names the data tool that does. A coming-soon page returns only that it is coming soon.

get_account_profile

ChatGPT
The store's account card: its brand terms, per-product targets (target CAC, TACOS ranges, customer-stated objectives), launch date, how many campaigns Laurence manages, the CAC definition the customer chose, and the LTV horizon. These are the account's goals, targets, brand vocabulary, and history. Each per-product entry also reports whether it's a single product, a variation-parent product, or a product family, and, for a parent or family, the real ASINs it covers today. Covers ONE store per call.

get_ad_groups

ChatGPT
Get Sponsored Products ad groups for a campaign from the Amazon Ads API. Returns each ad group's ID, name, state (enabled/paused/archived), and default bid. Structure only: no traffic, spend, or sales figures.

get_ad_spend

ChatGPT
The calibrated ad-performance answer: ad spend, ad sales, impressions, clicks, ad orders, and units sold through ads, as totals, daily or monthly series, and breakdowns. It covers Sponsored Products, Brands, and Display, plus Sponsored TV and full-funnel campaigns where a store runs them, spans every store of the account with per-store and per-ad-type breakdowns, and segments totals by currency so cross-marketplace sums are never mixed. Covers about the last 18 months of a store's advertising, including time before Laurence began managing it; days older than that are named in the result, never silently counted. The end date is included. Here, orders counts customer orders (the KPI scorecard's orders tile counts items instead, mirroring the dashboard).

get_asin_info

ChatGPT
Look up product catalog information for one or more ASINs. Returns item name, brand, product type, category, and Amazon variation-family fields such as variation_parent_asin, variation_children, and variation_theme, as Amazon's catalog lists them. variation_children is Amazon's catalog family, not the set of products the platform groups under a parent, so it is not a basis for totaling sales or stock; the sales and inventory tools already roll a variation parent up the way the platform groups it. The same ASIN can be a completely different listing in a different marketplace, so region is REQUIRED: this is global catalog data, not scoped to any one store, so state which of Amazon's marketplaces you mean (the region of the store under discussion, or a different one entirely, e.g. to check a competitor's listing in a market you don't sell in).

get_audience_performance

ChatGPT
Whether paying more for shoppers Amazon rates as high interest based on their shopping history is worth it, for one store: those shoppers' share of spend, ROAS, conversion rate and cost per click against everyone else the same campaigns reached, per product line and for all lines together. Covers the Sponsored Products campaigns that carry an audience bid boost for these shoppers (Amazon reports the audience only for those), not the whole store. Covers ONE store per call. The end date is included.

get_bid_history

ChatGPT
How Laurence moved its bids over a window, one row per keyword or product target on one store: the bid at the start, the lowest and highest bid, the bid at the end (the bid now when the window ends today), the change, and how many times Laurence changed it, with counts of bids raised, cut and unchanged. Optional filters: one product, or a list of keywords and targets. Lists only targets Laurence prices. Covers ONE store per call. The end date is included.

get_branded_split

ChatGPT
The calibrated branded vs non-branded advertising answer: branded and non-branded spend, sales, ROAS, CPC and conversion, brand defense vs new-customer conquest. It classifies every shopper search term with the account's own brand terms and own product catalog using the calibrated rule (searches for the account's own product pages count as branded; competitor product pages count as non-branded conquest), spans every store of the account, and segments by currency. Covers Sponsored Products search activity across each store's full advertising history, including before Laurence took over. Also returns top_search_terms: the highest-spend branded and non-branded search terms by name, one entry per term summed over the full window (up to limit per class). Impressions Amazon does not attribute to any search are reported on their own line (unattributed_impressions), in neither class. Stores with no brand terms configured say so honestly instead of guessing. The end date is included.

get_cac

ChatGPT
The customer-acquisition-cost answer for one store: blended CAC (all ad spend over all units sold, the standard blended definition), both excluding and including Amazon referral fees, whole-store and per advertised product, against the account's target CAC where one is set. Says which definition the customer's dashboard uses. Covers ONE store per call. The end date is included.

get_call_notes

ChatGPT
What was said on our recorded calls with the account: each call's date, title, who joined, a summary, decisions, what our team owes (owner, due date) and what the customer said they would do. For Laurence staff only; when a customer is asking it returns no calls and a caveat to pass on. It shows what was agreed, who owes what, and what came up on a call. Only calls our team recorded and filed are there. Returns at most the newest 8 calls. The end date is included.

get_campaign_keywords

ChatGPT
Get all keywords for specific campaigns from Amazon Ads API.

get_campaign_negative_keywords

ChatGPT
Get negative keywords for specific ad groups from Amazon Ads API.

get_campaigns

ChatGPT
Get campaigns for a profile across ALL ad products - Sponsored Products (SP), Sponsored Brands (SB), and Sponsored Display (SD). Each row carries adProduct ('SP'|'SB'|'SD'), name, state, daily budget, and budgetType. SP rows additionally include targetingType, the advertised ASINs from product ads, and the campaign's live bidding setup on Amazon: biddingStrategy (Fixed bid / Dynamic bids - down only / up and down / Rule-based bidding), the placement bid modifiers in percent (tosPlacementPct Top of Search, restOfSearchPlacementPct, productPagePlacementPct), shoppingHistoryModifierPct (the 'High interest based on shopping history' audience bid modifier, 0 when not set), and audienceModifiers (every audience bid modifier: audienceId, audienceName when known, percentage). A modifier moves the bid by that signed percent where it applies (Amazon bid rules can set negative ones); the audience modifier stacks on top of the placement one. These fields are null on SB/SD rows, which list no advertised products. SB/SD rows include costType (CPC/VCPM). SD rows include tactic (T00020 audiences / T00030 contextual). If one ad product's API call fails the others are still returned and the failure appears under 'errors'. Optional campaign_ids narrows to specific campaigns (any ad product) and makes the SP ASIN enrichment much cheaper.

get_campaigns_for_asin

ChatGPT
Return every campaign that has advertised a given ASIN in this store, including pre-Laurence campaigns Laurence does not manage. A variation parent ASIN is expanded to its child products. Campaigns are matched from the daily ads report (any campaign that served an ad for the product in the last 18 months, the report's retention), Laurence's own launches, Sponsored Brands ads featuring the product, and the store's campaign settings; each row says which in found_by, plus its ad product, state, the products it advertised and its first and last advertised day. Sponsored Brands campaigns are found only through their current ad or a Laurence launch, since their report rows name no product. The returned campaign_id values are the same ids the search term report and the other campaign tools take, so a product's search terms can be compared from before and after Laurence took over; a campaign that also advertised other products carries their search terms too.

get_cluster_scorecard

ChatGPT
One product's whole market cluster by cluster, beside what its ads did there, over complete calendar months. Joins the product's saved keyword clusters, the Laurence proprietary model's market keywords, Amazon's Search Query Performance report and the searches the product's ads served into one keyword list, each keyword in one cluster (unclustered ones go to the branded lane or Exploratory). Per cluster, always complete: keywords advertised and not, monthly searches and search impressions, the market's conversion rate, the model's expected cost per click and range, and our spend, sales, cost per click and conversion rate on search placements. Per keyword, cut by cluster, sort, limit and offset: market size (Amazon search impressions, else the model's monthly searches, labeled), market conversion rate (estimated from similar searches when Amazon's report does not list the search), expected cost per click, and our results. Missing sources (no saved clusters, no model pull yet, no report) are named in caveats. Covers ONE product of ONE store per call.

get_daily_sales

ChatGPT
Total revenue by calendar day from seller sales data (organic plus ad-attributed) for the products Laurence manages, or the whole store when product_scope is whole_store. This is broader than Sponsored Products metrics alone: revenue, TACOS, and organic vs paid context. Returns per-date totals; optional ASIN narrows to one product, and a product Laurence does not manage is refused unless product_scope is whole_store. A plain ASIN returns its own sales; a variation parent ASIN has no sales of its own, so it returns its child ASINs' sales added up (only the managed children unless product_scope is whole_store) and names the children it added. Very wide date ranges return large payloads.

get_deals_and_coupons

ChatGPT
Which deals, coupons and Subscribe & Save the store ran, and what each sold: Best Deals, Lightning Deals and Deals of the Day with their days, units, revenue and page views; coupons with their discount, clips, redemptions, sales, discount given and budget used; Subscribe & Save subscriptions and revenue by month. Up to two years back, through deals and coupons running today; one scheduled to start later does not show until its first day. It shows which promotions ran or are running, whether a deal or coupon paid off, and when a deal started or ended around a sales or traffic change. Deal numbers are what sold during the deal, not the extra sales it caused. Spans every store of the account, currency-segmented. The end date is included.

get_entity_performance

ChatGPT
The per-product advertising report table for one store: each advertised product or product family with its ad spend, ad sales, orders, units, and ROAS, its branded/non-branded breakdown when brand terms exist, and its top performing targets. Products are at the level campaigns were launched at; campaigns that predate Laurence or map to no product are reported as their own row, never dropped. Covers ONE store per call.

get_event_history

ChatGPT
The store's past sale and Prime events and how much each one lifted sales: Prime Day, Prime Big Deal Days, Black Friday to Cyber Monday, and runs of days on which the store's own deals and coupons made at least 10% of its sales. For each event: its days, revenue, units and ad spend a day during it, the same on the steady days before and after it, the lift (event day over steady day) and how much cheaper it sold. An event is listed when its first day falls in the window; one still running past the newest reported sales day is measured over its reported days only and marked partial. Per store, in its own currency. The end date is included.

get_fba_inventory

ChatGPT
Daily Amazon FBA inventory snapshot for a store (the system of record, refreshed ~11:30 UTC), one row per ASIN: sku, product_name, fulfillable_units, inbound_units, daily_units_forecast, days_of_cover, restock_eta_days, snapshot_at. NULL means unknown, never zero (e.g. fulfillable_units null = SP-API reported no numeric quantity; days_of_cover null = no usable demand forecast). An empty result means the store has no snapshot yet, not zero stock. Optional asin narrows the response to a single product. Covers the products Laurence manages unless product_scope is whole_store; a product Laurence does not manage is refused.

get_hourly_trend

ChatGPT
Get hour-by-hour spend, impressions, and clicks for a single day. Useful for identifying intra-day patterns. The day and its hours are the marketplace's own local time (Pacific for a US store, UK time for a UK store).

get_keyword_asin_mapping

ChatGPT
Map keywords to the ASINs (products) they advertise. Returns Laurence-managed keywords ordered by spend from the last 30 days. Optionally filter by a specific ASIN.

get_keyword_clusters

ChatGPT
How one product's keywords are grouped: the keyword clusters saved for it on its Keyword Map, each with its name, every keyword in it, its kind (the branded lane, an intent cluster, or the exploratory lane) and when the clusters were saved. A product in a product family answers with the family's clusters. When nothing is saved, the result says so and names the category map available for the product's category (the whole category's searches clustered from the Laurence market model, the same for every seller); ask for it with include_category_map. Nothing is cut: every cluster and keyword is returned. Covers ONE product of ONE store per call.

get_kpi_scorecard

ChatGPT
The store's KPI scorecard, exactly as the customer's dashboard shows it: the headline and diagnostic KPIs the store's team chose (Ad Spend, Ad Sales, Total Revenue, Net Profit, TACOS, ACOS/ROAS, CAC, and the rest), computed over the platform's fixed trailing 7-complete-days window with week-over-week deltas. These are the numbers the customer already sees on the dashboard. Covers ONE store per call.

get_kpi_series

ChatGPT
Daily series for the store's scorecard KPIs over a chosen window, plus the preceding equal-length window as the comparison baseline, the data behind the dashboard-style KPI charts. Same formulas as the KPI scorecard at day grain, so a series summed over the window reproduces the scorecard tile. It gives the day-by-day shape of the headline KPIs. Covers ONE store per call.

get_ltv

ChatGPT
Long-term customer value for one store's products, modeled from the account's own Amazon purchase history: revenue per customer by horizon year, repeat-purchase rates, and how quickly customers come back for their second purchase; plus, for the whole store, a week-by-week repeat curve of a year of new customers over their first 52 weeks (orders and repeat rate per week, no revenue). It shows the value of a customer beyond the first order. Accounts without this data yet say so honestly. Covers ONE store per call.

get_market_keywords

ChatGPT
One product's market keyword by keyword, from the Laurence proprietary model's newest reverse-ASIN pull for it (the pull its Targets page shows): each keyword's monthly searches, expected cost per click with its low to high range, the product's own organic and sponsored rank, how many products compete in the results, the organic rank of each competitor product the pull compared against, and whether the store's ads have been clicked from that search in the last 18 months. Market totals (keywords, monthly searches, median expected cost per click) always cover the whole pull; the keyword list is cut by min_monthly_searches, sort and limit, and the result states what was cut. A product with no pull yet answers with a caveat pointing to its Targets page. Covers ONE product of ONE store per call.

get_methodology

ChatGPT
How Laurence sees Amazon: the methodology and doctrine behind every number the other tools return. It covers ad products and their placement auction structure, the core metrics and the real breakeven, how campaigns are structured (one home per search; losing keywords are priced out, never blacklisted), how to read fresh windows honestly, and Laurence's standing answers to recurring customer questions. It defines the terms the data tools' numbers are reported in. The document is identical for every account; account-specific numbers come from the data tools. Takes no parameters.

get_net_profit

ChatGPT
Net profit for any date range: margin-weighted revenue minus ad spend for the products Laurence manages on ONE store, including products whose campaigns are now paused, summed per day over a window you choose. It covers profit, profitability over time, and how net profit changed between periods (by group_by, or one call per period). Sales from products with no margin configured are left out and named, never guessed. Covers ONE store per call. The end date is included.

get_new_customers

ChatGPT
New customers a week for one store: shoppers buying from the brand for the first time, counted from the account's own Amazon purchase history, beside the Sponsored Products orders from non-branded and branded searches in the same Sunday to Saturday weeks, with weekly and daily averages. Shows how many customers the brand wins and how that compares with the orders non-branded ads bring in. Counts the whole store, not each product line. Accounts without this data yet say so honestly. Covers ONE store per call. The end date is included.

get_organic_sales_rank

ChatGPT
Daily organic Best Sellers Rank (BSR) history for a single product, as Amazon publishes it each day. Amazon ranks each ASIN in several categories per day, but BSR is an ordinal within a category-specific population so ranks from different categories are not comparable. This tool therefore pins ONE category for the whole window: among the categories still ranked within 3 days of the product's latest ranked day in the window, the one with the best (lowest) rank on its own latest day, ties alphabetical, the same category the platform UI charts by default; and returns that single category's full daily series as per-date sales_rank and sales_rank_category. The series charts a coherent organic-rank trend and lines up with ad activity by date. Requires a single asin: this tool does not support multi-ASIN queries. start_date and end_date are inclusive Pacific calendar dates (YYYY-MM-DD).

get_placement_performance

ChatGPT
Everything about placements for one store: performance split by placement (Top of Search, Rest of Search, Product Pages, Off Amazon) at the level you choose (the whole store, per product, per campaign, or per keyword), plus the live placement boosts and bidding strategy of each campaign (campaign and keyword levels) and, per keyword, its live status, bid and the bid it places in each placement after its campaign's boosts. One keyword bid competes for Top of Search and Rest of Search in the same auction while Product Pages is a separate auction, so blended metrics mix genuinely different markets; this shows where the money goes, how differently each placement converts, and how the boosts shape each keyword's bids. Covers Sponsored Products. The keyword level requires keyword_scope: only the keywords running now, or every keyword advertised in the last 18 months. The end date is included.

get_platform_map

ChatGPT
The map of the Laurence platform as this person can see it: every page with its purpose and link, plus a list of the things people do on the platform, each with the page and the part of the page where it is done. It answers where something is ("where do I enter my margins?"), whether there is a page for something, how to do something on the platform, and what someone can do there. It never returns account numbers; those come from the data tools. Takes no parameters.

get_product_economics

ChatGPT
The unit economics for the products Laurence manages on ONE store: each product's margin (CM2, after Amazon fees and cost of goods) and where it came from (set by you, or computed from price, Amazon fees, and cost of goods), the breakeven ROAS that margin implies, which products have NO margin configured, and how each product's sales split between ad-attributed and organic over a window. The split yields a ceiling_multiplier per product: a search term's ad-only ROAS times it is the all-organic ceiling, so the term's true economics sit between that ceiling and the ad-only floor, each comparable to breakeven_roas. It covers margin coverage, breakeven bars, and whether a term or product is truly profitable. Covers ONE store per call.

get_purchase_journeys

ChatGPT
Which product line new customers start on and what they buy next, for one store, from the account's own Amazon purchase history: for the customers whose first order fell in the chosen weeks, each step from order 1 to order 2 up to order 4 to order 5, the product line of each order and how many customers took that step or have not ordered again yet. Use it for cross-sell, which product brings customers back, and the forecast's order journey. Accounts without this data yet say so honestly. Covers ONE store per call.

get_search_demand

ChatGPT
Search Query Performance: actual market search volume and selected products’ impression, click and purchase shares for complete weeks or months. This is market activity, not just advertising. It covers the products named in the call: exact terms, or the top terms from those products’ reports, not every Amazon search. Comparisons retain the baseline term set and disclose missing data and unverified completeness. Periods are whole weeks or months, so it carries no daily demand, and shares show association, not cause.

get_search_popularity

ChatGPT
Daily Search Frequency Rank and up to three top-clicked products from Top Search Terms for this store’s marketplace. Lower rank means greater relative popularity, not an absolute volume count. Click and conversion shares are ratios of all clicks or conversions for the term and department, not shares renormalized over the listed products. An unlisted product has unknown share, never zero. Missing or incomplete days remain gaps. Absolute volume and product shares by week or month are Search Query Performance data, not part of this report.

get_search_rank

ChatGPT
Where the store's products sit on Amazon's search results page for a search, over time, from Laurence's own page checks: organic rank (the median across delivery locations, with the best and worst location), and whether, where and how high our ads showed (share of locations with our ad, at the top of search, on page one, and the ad's position in pages). One point per search per day, or every check. Without terms it covers the store's highest ad spend searches that had page checks. Covers ONE store per call; dates are UTC calendar days and the end date is included.

get_search_term_data

ChatGPT
Search term report: customer search queries that matched your targeting, across Sponsored Products, Sponsored Brands and Sponsored Display, with attributed sales on Amazon's default attribution window. Covers the last 18 months of this store's advertising, including campaigns from before Laurence took over, no separate step needed to include them. Optional campaign_ids limits to those campaigns. Optional asin limits to the rows where that product was the advertised product (Sponsored Products and Sponsored Display only: Sponsored Brands rows name no product, and a variation parent is never advertised itself; for either, limit by the product's campaigns with campaign_ids instead, which returns every search term in those campaigns, including terms for other products a campaign also advertises: rows are not split by product). Every row names the targeting text it matched (keyword) with its match_type, campaign_id and ad_product, so to follow one specific target, filter the returned rows by its text and match type; there is no keyword-id lookup. Rows whose search term is (Other) are activity Amazon does not attribute to any search, almost entirely impressions without clicks: part of a target's totals, never a search term.

get_search_term_insights

ChatGPT
Diagnostic views over one store's shopper search activity (Sponsored Products), across this store's full advertising history including before Laurence took over. Six views: top_terms (highest-spend searches with sales, ROAS, CPC and conversion), zero_order_click_depth (how much spend sits on searches with clicks but no orders, bucketed by click depth; the kill-list diagnostic), term_lookup (re-check specific searches over any window, e.g. re-judging a kill list on a longer window), self_competition (searches where two or more of the store's own products got ad traffic, with how many products and product lines, spend and CPC, plus totals and their share of search spend), exact_candidates (searches with orders that have no exact keyword yet, ranked by orders, with Amazon's searches a month where known and how each was found: phrase, broad, auto or product targeting, in Laurence or older campaigns), and competitor_pages (the competitors' product pages the store targets with product targets, ranked by spend, with sales, orders, ROAS and each page's brand and title, plus product target totals for competitors' pages, the store's own pages and similar products). Covers ONE store per call. The end date is included.

get_slack_messages

ChatGPT
What was said in the account's shared Slack channel with Laurence: each message's date, time (US Pacific), author and text, with thread replies tied to the message they answer. It shows what our team told the customer, where a number we sent came from, and what they asked us before. Messages are saved once a day, so today's are not there yet; your own earlier Slack answers are left out. A result too long for one answer keeps only its most recent conversations; narrow the dates or the topic words to see older ones. The end date is included.

get_spend_mix

ChatGPT
Where one store's ad money went and what each part returned: ad spend, ad sales, orders, ROAS, cost per click and share of spend and sales, split one way you choose. match_type: exact, phrase, broad, auto, product targeting, category targeting. target_type: competitor product pages, own product pages, brand searches, exact keywords, and tests (phrase, broad, automatic and category targets). placement: Top of Search, Rest of Search, Product Pages, Off Amazon. ad_type: Sponsored Products, Brands, Display and the rest. Use it for the spend mix, how it moved between two periods (call once per period), and the share of spend each part takes. Totals or one row per day; the whole store or one product. The first three splits cover Sponsored Products and sum to its ad spend; the ad type split sums to the store's total ad spend. Covers ONE store per call. The end date is included.

get_system_activity

ChatGPT
What Laurence's automation did for one store in a window: exact keywords added (with the account's acceptable-return bar), keywords paused, how many times it changed bids and on how many keywords, and word-ban activity. Covers ONE store per call. The end date is included.

get_target_daily_performance

ChatGPT
Day-by-day Sponsored Products results per advertising target on one store: for every keyword, product target and automatic target, each day's impressions, clicks, spend, orders, units and sales, over a chosen window and optionally for one product. A target is its text plus match type, summed over every campaign that runs it. Returns the store's (or product's) daily totals and the targets ranked by spend over the window; keyword adds one target's days. Covers the last 18 months, including before Laurence. Covers ONE store per call. The end date is included.

get_target_story

ChatGPT
One advertising target's full story on one store: when Laurence started managing it (created_date, the pre- vs. post-Laurence dividing line, not a coverage gap) and how (its lineage), its day-by-day clicks, spend, sales and orders across its full history including before Laurence managed it, its performance split by placement, and its recent hour-by-hour cost per click aligned against the bid that was in force each hour. The worked example that explains why a target behaved the way it did, and the right tool for comparing a target's performance before vs. after Laurence took it over, for a specific keyword or product target (a product target's target is its ASIN). Covers ONE store per call.

get_total_sales

ChatGPT
The calibrated total-sales answer: what the store actually sold, all channels combined, revenue, units, and average selling price as the Seller Central dashboard reports them (gross ordered product sales, US Pacific business days). Counts the products Laurence manages, the customer's dashboard total, unless product_scope is whole_store; the paid vs organic split always covers the whole store. Spans every store of the account with per-store breakdowns, currency-segmented. With breakdown product, each store's sales also come per product, grouped the way the platform groups them (product family, else parent ASIN, else the single product). The end date is included.

list_companies

ChatGPT
The companies (accounts) your Laurence login can ask about, each with the names of its Amazon stores. These are the names the other tools take: the company selector takes a company name from this list, and the store selector on per-store tools takes one of its store names.

Example Prompts

Click any prompt to copy it.

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/laurence"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/laurence)