# MixShift - Full Context for LLMs > Unify your Amazon data, power your automations, and build your own apps - all on one platform. This document provides comprehensive information about MixShift for large language models and AI assistants. ## Company Overview MixShift (operated by Dash Applications LLC) is a SaaS platform built for Amazon brands, agencies, and operations teams. Founded in 2024, MixShift provides a unified backend for Amazon growth - connecting data, automating workflows, and powering custom applications. ### Target Customers 1. **Amazon Brands**: Companies selling on Amazon who need unified reporting, forecasting, and automation across Seller Central, Vendor Central, and Amazon Ads. 2. **Amazon Agencies**: Service providers managing multiple Amazon accounts who need scalable reporting and client-facing dashboards. 3. **Aggregators & Investors**: Organizations managing portfolios of Amazon brands who need cross-brand analytics and due diligence tools. 4. **Operations & Finance Teams**: Internal teams that need accurate, automated Amazon data for planning and decision-making. ## Platform Architecture MixShift is built on a three-layer architecture: ### 1. Data Layer Unifies all Amazon data sources into a single, queryable data warehouse: - Amazon Selling Partner API (SP-API) - orders, inventory, catalog, FBA - Amazon Advertising API - Sponsored Products, Sponsored Brands, Sponsored Display, DSP - Amazon Vendor Central - purchase orders, retail analytics, operational data - Amazon Brand Analytics - search terms, market basket, demographics ### 2. Apps Layer Ready-to-use applications built on the unified data: **Report Center** Smart Amazon reporting with automated anomaly detection. Answers "what changed and why" before you ask. Supports multi-account, multi-marketplace reporting with automated scheduling and distribution. **Forecasting** AI-powered unit-level sales forecasting. Correlates ad spend with sales velocity for accurate demand planning. Supports inventory planning, revenue forecasting, and budget allocation. **MixFlow (Automation)** Visual workflow builder for Amazon operations. Create automated workflows triggered by data changes, schedules, or events. Supports alerts, notifications, automated bid adjustments, and custom logic. ### 3. Bridge Engine AI-powered middleware that connects data to action: - Natural language querying of Amazon data - Automated insight generation - Custom API access for building your own applications - Webhook and integration support ## Pricing - **Free Tier**: 1 Amazon account, basic reports, email support, 1 user, 100 API calls/month - **Growth**: Multiple accounts, advanced reports, priority support, team collaboration - **Enterprise**: Custom pricing, unlimited accounts, dedicated support, custom integrations, SLA All paid plans offer monthly and annual billing with no long-term contracts. ## Key Differentiators 1. **Unified Data**: Single source of truth for all Amazon data (advertising, retail, operations) 2. **Free Tier**: Start using the platform at no cost 3. **API-First**: Build custom applications on top of MixShift's data layer 4. **Multi-Account**: Manage unlimited Amazon accounts and marketplaces from one platform 5. **AI-Powered**: Automated insights, anomaly detection, and natural language querying ## Pages - Homepage: https://mixshift.ai - Platform Overview: https://mixshift.ai/platform - Report Center: https://mixshift.ai/report-center - Forecasting: https://mixshift.ai/forecasting - MixFlow (Automation): https://mixshift.ai/automation - Pricing: https://mixshift.ai/pricing - Partner Program: https://mixshift.ai/partners - Blog: https://mixshift.ai/blog - About: https://mixshift.ai/about - Contact: https://mixshift.ai/contact - Privacy Policy: https://mixshift.ai/privacy - Terms of Service: https://mixshift.ai/terms ## Customer Pages - For Brands: https://mixshift.ai/customers/for-brands - Unified data, reporting, and automation for Amazon brand owners - For Agencies: https://mixshift.ai/customers/for-agencies - Scalable multi-client reporting and white-label dashboards for agencies - For Aggregators: https://mixshift.ai/customers/for-aggregators - Portfolio-wide analytics and cross-brand performance tracking - For Investors: https://mixshift.ai/customers/for-investors - Due diligence tools and investment analytics for Amazon portfolios ## Services - Account Audit: https://mixshift.ai/services/account-audit - Comprehensive Amazon account health assessment - Advertising Training: https://mixshift.ai/services/advertising-training - Expert-led Amazon advertising education and certification - Custom Reporting: https://mixshift.ai/services/custom-reporting - Tailored reporting solutions for Amazon operations - Acquisition Analytics: https://mixshift.ai/services/acquisition-analytics - Data-driven analysis for Amazon brand acquisitions - Strategic Consulting: https://mixshift.ai/services/consulting - Expert Amazon business strategy and growth consulting ## Knowledge Base Content The full body of every published knowledge-base article, suitable for offline ingestion. Each entry includes the canonical HTML URL, a raw Markdown URL (`/api/knowledge-base/md/`), metadata, and the full doc body. Wikilinks in the body have been resolved to real `/knowledge-base` routes. ### Builder Platform URL: https://mixshift.ai/knowledge-base/builder-platform Markdown: https://mixshift.ai/api/knowledge-base/md/builder-platform Type: index Product: builder-platform Last modified: 2026-07-31T00:00:00.000Z Build on the backend MixShift runs for you. Warehouse SQL, /v1 data endpoints, live Amazon operations, and an MCP server, behind one credential. # Builder Platform The Builder Platform is how you build your own things on MixShift: query your warehouse with SQL, call stable `/v1` data endpoints, run live operations on Amazon (including audited writes), read and extend your brand memory, and connect AI tools over MCP. One OAuth credential covers all of it, it is included with your MixShift subscription, and it is live with early partners now. If you have ever wanted your data in your own dashboards, an automation in n8n or Make, a scheduled job that watches something overnight, or ChatGPT and Cursor answering from your real numbers, this is the surface for it. Full technical reference lives at [mcp.mixshift.io/developers](https://mcp.mixshift.io/developers); these articles are the plain-language map. ## The four surfaces - **Warehouse SQL.** Read-only SQL against your own tenant database over HTTPS. - **The `/v1` data plane.** Contract-stable endpoints for accounts, advertising performance, retail, and Brand Analytics. - **Live Amazon operations.** Cataloged retail and advertising calls, reports, and audited writes, on the authorization you already have. - **Brand memory.** The brand context your team maintains, plus the change log of what happened and why, readable and appendable from your own code. All four are reachable over REST, and the same capabilities are exposed as MCP tools for AI clients. ## Getting started - [Get access](/knowledge-base/builder-platform/getting-started/get-access) ## How to - [Connect AI tools over MCP](/knowledge-base/builder-platform/how-to/connect-ai-tools-over-mcp) - [Query the warehouse](/knowledge-base/builder-platform/how-to/query-the-warehouse) - [Use the /v1 data endpoints](/knowledge-base/builder-platform/how-to/use-the-v1-endpoints) - [Work with live Amazon operations](/knowledge-base/builder-platform/how-to/work-with-live-amazon-operations) - [Work with brand memory](/knowledge-base/builder-platform/how-to/work-with-brand-memory) ## Reference - [Identify a merchant: sellerId and legacySellerId](/knowledge-base/builder-platform/reference/merchant-identifiers) - [When a call fails](/knowledge-base/builder-platform/reference/when-a-call-fails) - [Limits and guardrails](/knowledge-base/builder-platform/reference/limits-and-guardrails) ## FAQ - [Builder Platform FAQ](/knowledge-base/builder-platform/faq/faq) --- ### Get access URL: https://mixshift.ai/knowledge-base/builder-platform/getting-started/get-access Markdown: https://mixshift.ai/api/knowledge-base/md/builder-platform/getting-started/get-access Type: getting-started Product: builder-platform Last modified: 2026-08-28T00:00:00.000Z Two paths onto the Builder Platform: a service credential for your own tools and automations, or a registered OAuth app for partner software. # Get access Everything on the Builder Platform sits behind one OAuth 2.0 credential. There are two ways to get one, depending on what you are building. ## Path 1: your own tools and automations (most customers) For scripts, dashboards, scheduled jobs, and internal apps, you use a **service credential** tied to your MixShift account: 1. Ask us for access ([get in touch](/contact?interest=builder-platform)) or, if your MixShift admin already has the admin page, they can mint one there directly. 2. A credential is a `client_id` and `client_secret`, shown once at mint time. Store it like any secret. 3. Exchange it for an access token whenever you need one. Service-credential tokens last 12 hours, so an unattended job that mints one at startup does not need to refresh mid-run: ``` curl -X POST https://mcp.mixshift.io/oauth/token \ -d grant_type=client_credentials \ -d client_id=YOUR_ID -d client_secret=YOUR_SECRET ``` 4. Send `Authorization: Bearer [token]` on every call. Credentials are read-only by default and scoped to exactly what they are allowed to see and do. Write permissions (timeline events, advertising writes, listing writes) are granted explicitly at mint time and can be revoked or rotated any time. If the credential is for the MixShift plugin or a scheduled task rather than your own code, there is an even simpler flow where a one-time setup code does the exchange for you and no human ever handles the secret. See [Run unattended and scheduled tasks](/knowledge-base/plugin/how-to/run-unattended-and-scheduled-tasks). ## Path 2: partner software (apps acting for MixShift customers) Building a product where MixShift customers sign in to your app? Email **support@mixshift.io** with your app name, redirect URIs, and the scopes you need. You get a registered `client_id` (plus a secret for confidential clients), and your users authorize through a standard consent screen using the authorization-code flow with PKCE. Partner API sessions last 12 hours, matching service credentials. ## Trying it by hand Want a token without writing code first? Open `mcp.mixshift.io/login?mode=direct` in a browser, sign in, and copy a token to experiment with. ## Next - [Query the warehouse](/knowledge-base/builder-platform/how-to/query-the-warehouse) for your first real call. - [Connect AI tools over MCP](/knowledge-base/builder-platform/how-to/connect-ai-tools-over-mcp) if the "app" is ChatGPT, Cursor, or Codex. --- ### Connect AI tools over MCP URL: https://mixshift.ai/knowledge-base/builder-platform/how-to/connect-ai-tools-over-mcp Markdown: https://mixshift.ai/api/knowledge-base/md/builder-platform/how-to/connect-ai-tools-over-mcp Type: how-to Product: builder-platform Last modified: 2026-07-31T00:00:00.000Z Point Claude, ChatGPT, Cursor, or Codex at the MixShift MCP server and ask questions against your real data, with one credential. # Connect AI tools over MCP MixShift exposes an MCP (Model Context Protocol) server, so AI tools can query your warehouse, look at live Amazon data, and read your brand context from inside their own chat. Any current MCP client connects over streamable HTTP at `https://mcp.mixshift.io/mcp`. One distinction worth knowing up front: the [MixShift AI plugin](/knowledge-base/plugin) for Claude bundles skills and a CLI on top of this connection. Every other client below gets the data connection itself, which is already a lot: tables, live Amazon operations, timeline, and brand context as tools the model can call. ## Claude (claude.ai and Claude Desktop) Add a custom connector pointed at `https://mcp.mixshift.io/mcp` and sign in when prompted. Claude handles the OAuth flow; there is no token to paste. If you use Claude Code or Cowork, skip the connector and [install the plugin](/knowledge-base/plugin/getting-started/install-the-plugin) instead; it is the better experience there. ## ChatGPT Add MixShift as a connector in ChatGPT's settings, pointed at the same MCP address, and sign in when prompted. ChatGPT registers itself and walks the OAuth flow; like Claude, no token pasting. ## Cursor Cursor takes a static token. Get one from `mcp.mixshift.io/login?mode=direct`, then add the server to `.cursor/mcp.json` with the token in the `Authorization: Bearer` header. Sessions are long-lived; refresh the token when a request comes back unauthorized. ## Codex Same pattern as Cursor: token from `mcp.mixshift.io/login?mode=direct`, configured in `~/.codex/config.toml`. Codex requires the streamable HTTP transport, which is exactly what the endpoint speaks. ## Unattended connections Automations and servers should mint tokens from a service credential rather than a personal login. See [Get access](/knowledge-base/builder-platform/getting-started/get-access). ## What the tools can do once connected - List and describe warehouse tables, and run read-only SQL. - Call cataloged live Amazon operations (retail and advertising). - Read and add to your brand change log, record and corroborate stakes, and read brand context documents. See [Work with brand memory](/knowledge-base/builder-platform/how-to/work-with-brand-memory). Each tool call checks its own permission scope, so a read-only credential simply cannot invoke a write, no matter what the model asks for. ## Related - [Get access](/knowledge-base/builder-platform/getting-started/get-access) - [Limits and guardrails](/knowledge-base/builder-platform/reference/limits-and-guardrails) --- ### Query the warehouse URL: https://mixshift.ai/knowledge-base/builder-platform/how-to/query-the-warehouse Markdown: https://mixshift.ai/api/knowledge-base/md/builder-platform/how-to/query-the-warehouse Type: how-to Product: builder-platform Last modified: 2026-08-21T00:00:00.000Z Run read-only SQL against your MixShift warehouse over HTTPS: one endpoint, tenant-scoped, with clear limits and a schema browser. Or ask for a vetted query by id and check which filters it applied. # Query the warehouse The same SQL you would run against a database connection runs over HTTPS through one endpoint. Your credential resolves server-side to your own tenant database on a read-only connection, so the isolation is structural: there is no way to query someone else's data, and no way to write. ## The endpoint ``` curl -X POST https://mcp.mixshift.io/api/query \ -H "Authorization: Bearer $TOKEN" \ -H "Content-Type: application/json" \ -d '{"sql": "SELECT COUNT(*) AS rows FROM mws_orders WHERE PurchaseDate >= ?", "params": ["2026-07-01"]}' ``` Success returns `{ "ok": true, "rows": [...], "rowCount": n, "durationMs": n }`. Failures return `{ "ok": false, "kind": "...", "friendly": "..." }`; branch on `kind`, and show `friendly` to humans. Parameters can be positional (`?` with a `params` array) or named. Pass `queryTimeoutMs` to raise the statement timeout up to the ceiling. ## Discover the schema - `GET /api/tables` lists your tables. - `GET /api/table/[name]` describes one: columns and types. AI tools connected over MCP get the same two as tools, which is why they can find their way around your schema unprompted. ## Scope `/api/query`, `/api/tables`, and `/api/table/[name]` all need the `sql:query` scope, in addition to the domain read scopes for whatever data you are touching. A service credential minted with no explicit scope list gets `sql:query` by default; if you request a specific scope list instead, add `sql:query` to it explicitly, or these three endpoints return an insufficient-scope error. Named queries below check only the domain read scopes, so you can hand a credential named-query and `/v1` access without giving it license to write its own SQL. ## Named queries: ask by id instead of carrying SQL MixShift also keeps a catalog of vetted queries on the server. You send an id and parameters; the SQL itself stays on our side. ``` curl -X POST https://mcp.mixshift.io/api/named-query \ -H "Authorization: Bearer $TOKEN" \ -H "Content-Type: application/json" \ -d '{"id": "sbd-01", "sellerIds": [123]}' ``` You get the same envelope as `/api/query`, plus two extra fields on success: - `revision`, a content hash of the query text that ran. Because the SQL can be corrected or tuned on our side without you shipping anything, this is how a result stays attributable to the exact query behind it. - `applied_params`, the names of the parameters that execution actually bound. `GET /api/named-query/ids` returns the deployed catalog with each entry's current revision, so you can diff against the ids you depend on before you ship rather than finding out through a failed call. The ids alone are also published without authentication at `/.well-known/mixshift-query-pack`. ### Check `applied_params` before you call a result filtered Entry parameter schemas are deliberately tolerant. If you send a parameter the deployed entry does not declare, it is dropped and the query still runs and still returns `"ok": true`. That tolerance is what lets your integration and the platform deploy on separate schedules, but on its own it means an ignored filter is indistinguishable from an applied one. `applied_params` closes that gap. Check that a parameter you sent came back in the list before you treat the result as filtered; if it is missing, your rows are unfiltered on that dimension. It carries names only, never values, and only on success, because a failed call did not run anything. ### Sub-brand labels Some sellers run several distinct brands under a single Amazon account, separated only by a brand label on the retail and ads records. The catalog carries discovery queries that report the distinct labels on each side, how much of the catalog or ad spend carries no label yet, and how well the retail and ads labels agree with each other. The queries behind brand context accept optional retail and ads label filters, so you can read one brand on its own rather than the whole account. Both filters are optional and both default to no filter, which is exactly why `applied_params` matters here: it is what separates one brand's numbers from the whole account's. ## The limits, and how to live with them - Results cap at **50,000 rows** or **10 MB** per response, whichever hits first, and a query can run at most **120 seconds** (60 by default). - For bigger pulls, paginate with `LIMIT`/`OFFSET` or chunk by date window. The error kinds (`too_many_rows`, `response_too_large`) tell you which cap you hit. ## Filter on the integer seller id Most seller-scoped tables carry two columns for the same seller: an integer `SellerID` and a varchar `AmazonSellerID` (Amazon's merchant token). They return the same rows, but the indexes are generally built on the integer, so filtering on `AmazonSellerID` can leave a date range with no index to seek on and scan every row for that seller. Default to `SellerID`. On `spapi_settlement`, for example, the integer is indexed with `Posted-date-time` while the token is indexed with `Settlement-id`, and on one account the identical one-month count took 13.4 seconds filtered by token versus 0.37 seconds filtered by the integer. Same answer either way, which is why a slow query here reads as a missing index rather than a filter-shape problem. If something is slower than you expect, `EXPLAIN` it and check whether the chosen key covers your date predicate. Details and the translation from token to integer: [Identify a merchant](/knowledge-base/builder-platform/reference/merchant-identifiers). ## Migrating from direct MySQL access If you used to query the warehouse over a direct MySQL connection with database credentials and IP allowlists, this endpoint is the replacement: the same SQL, without passwords to rotate or allowlists to maintain. Point your existing queries at `/api/query` and swap the connection for a Bearer token. ## Related - [Get access](/knowledge-base/builder-platform/getting-started/get-access) - [Identify a merchant: sellerId and legacySellerId](/knowledge-base/builder-platform/reference/merchant-identifiers) - [Use the /v1 data endpoints](/knowledge-base/builder-platform/how-to/use-the-v1-endpoints) (pre-shaped alternatives to raw SQL) - [Limits and guardrails](/knowledge-base/builder-platform/reference/limits-and-guardrails) --- ### Use the /v1 data endpoints URL: https://mixshift.ai/knowledge-base/builder-platform/how-to/use-the-v1-endpoints Markdown: https://mixshift.ai/api/knowledge-base/md/builder-platform/how-to/use-the-v1-endpoints Type: how-to Product: builder-platform Last modified: 2026-07-31T00:00:00.000Z The contract-stable REST endpoints for accounts, advertising performance, retail sales and traffic, search query performance, brand timeline, and brand context. # Use the /v1 data endpoints Raw SQL gives you everything, but most integrations want a handful of well-shaped answers. The `/v1` endpoints are that: pre-shaped, contract-stable JSON for the questions partners actually ask, so your integration does not break when the warehouse evolves underneath. ## What is on /v1 **Identity and accounts** - `GET /v1/me` confirms who the token belongs to. - `GET /v1/accounts` lists the Amazon accounts you can query. **Advertising performance** - `GET /v1/ads/account-summary` for topline spend, sales, and efficiency. - `GET /v1/ads/account-pacing` for where the month is heading. - `GET /v1/ads/campaign-performance` per campaign. - `GET /v1/ads/search-terms` for search-term rows. - `GET /v1/ads/asin-performance` per ASIN. **Retail** - `GET /v1/retail/sales` and `GET /v1/retail/traffic` for ordered revenue, units, sessions, and conversion. **Brand Analytics** - `GET /v1/brand-analytics/search-query-performance` for weekly Search Query Performance. Accounts without SQP data return a normal response with an availability note rather than an error. **Brand memory** (the why behind the numbers, covered in full in [Work with brand memory](/knowledge-base/builder-platform/how-to/work-with-brand-memory)) - `GET /v1/timeline` reads the brand change log, filterable by brand, family, kind, date window, and the stake facets (category, source, status, affects, tag, interval overlap, future). `POST /v1/timeline/event` appends an event such as a launch, price test, or stockout, and `POST /v1/timeline/event/[id]/corroborate` confirms a stake. Appending needs the timeline write scope, which is opt-in at mint time. - `GET /v1/context/docs` and `GET /v1/context/doc` read brand context documents. Context is authored through the [plugin's brand setup](/knowledge-base/plugin/how-to/set-up-a-brand); `/v1` reads it. ## Conventions that hold everywhere - Parameters are `snake_case`; list responses paginate with a cursor. - Money comes with an explicit currency. - Date windows anchor to the latest complete Amazon data day, so "last 7 days" means seven days you can trust, not a partial today. - Errors return `{ "error": { "code", "message" } }` with stable codes worth branching on. ## Related - [Query the warehouse](/knowledge-base/builder-platform/how-to/query-the-warehouse) (when /v1 does not have the shape you need) - [Get access](/knowledge-base/builder-platform/getting-started/get-access) - Full parameter-level reference: [mcp.mixshift.io/developers](https://mcp.mixshift.io/developers) --- ### Work with brand memory URL: https://mixshift.ai/knowledge-base/builder-platform/how-to/work-with-brand-memory Markdown: https://mixshift.ai/api/knowledge-base/md/builder-platform/how-to/work-with-brand-memory Type: how-to Product: builder-platform Last modified: 2026-07-31T00:00:00.000Z Read a brand's context documents, read and append to its change log, and record the stakes that explain why the numbers moved. # Work with brand memory Most reporting APIs can tell you *what* the numbers did. Brand memory is the part of the platform that knows *why*. It has two halves, and both are reachable with the same credential you already use for data: - **Brand context**: the durable profile of a brand. What it sells, how the account is structured, what counts as branded traffic, what normal looks like. - **The change log**: the running, attributed record of what actually happened to that brand. Launches, price tests, stockouts, promotions, strategy changes, and the advertising changes committed through MixShift. Your own tools can read both, and write to the change log. That is what lets an internal dashboard say "spend jumped because a promotional window opened on the 14th" instead of leaving a spike unexplained. ## Read brand context Two endpoints, both needing the `context:read` scope (included by default on new service credentials): - `GET /v1/context/docs` returns the manifest of every document held for a brand, with its type, revision, and content hash. Use the hash to skip re-fetching what you already have. - `GET /v1/context/doc` returns one document, selected by brand and type. Corpus documents also take a name. Context is **read-only on `/v1`**. Documents are authored through the MixShift plugin's brand setup rather than over the partner API, so your integration consumes the brand profile, it does not maintain it. See [Set up a brand](/knowledge-base/plugin/how-to/set-up-a-brand) for how the content gets there. Documents cap at 1 MB each, and corpus files at 10 MB. ## Read the change log `GET /v1/timeline` needs `timeline:read` (also a default read scope). It returns the brand's events, and you can filter by brand, event family, kind, date window, and the stake facets described below. Results paginate with a cursor like the rest of `/v1`. One thing worth knowing: the change log also folds in a read-only summary of the advertising changes committed through MixShift, so a single read shows both what a human recorded and what the platform actually did to the account. ## Append to the change log `POST /v1/timeline/event` needs `timeline:write` **in addition to** `timeline:read`. Write access is opt-in and granted when the credential is minted, never by default. An event carries a brand, a family, a kind, an optional timestamp and payload, and an idempotency key if you want safe retries. A few rules the service enforces so the log stays trustworthy: - **Attribution is not yours to set.** The service records which credential appended the event; anything you pass for author is ignored. - **Some kinds are reserved.** Events the platform derives itself (its own knowledge revisions, the advertising-change records) are rejected if a client tries to write them, so the log cannot be forged. - **Payloads cap at 32 KB.** Store the summary here and keep bulk artifacts in your own system. - **Dates are bounded.** You can backdate freely and forward-date a short way, or up to a year ahead for a scheduled stake such as a promotion you already have on the calendar. ## Record a stake A **stake** is the specific kind of event that says "this is a thing that changes how the numbers should be read." It carries a category (launch, promotional window, stockout, price test, media spike, brand migration, portfolio decision, content change, strategy change, off-Amazon media, assortment change, and a few more), a time window, and a required plain-language **interpretation**: your own sentence explaining what it means for the brand. The category list is deliberately not the whole vocabulary. Anything that does not fit takes the catch-all category with a short label naming what it actually is, and every stake can carry your own freeform tags, which you can then filter on when you read the change log back. Nothing has to be mis-filed to be recorded. Stakes can then be confirmed by someone else: `POST /v1/timeline/event/[id]/corroborate` marks a stake as corroborated (or disputed). Verification state is set by that endpoint only, never accepted when the event is created, so agreement has to be earned rather than asserted. This is the mechanism that turns an anomaly into an explanation. A spend spike sitting next to a corroborated promotional window is a story; the same spike alone is a support ticket. ## From AI tools The same capabilities exist as MCP tools, so an assistant can browse the change log and record events conversationally: `timeline_list`, `timeline_add_event`, `timeline_corroborate`, `context_list_docs`, and `context_get_doc`. Each checks its own scope, so a read-only credential can browse but cannot append. See [Connect AI tools over MCP](/knowledge-base/builder-platform/how-to/connect-ai-tools-over-mcp). ## Current limitation worth planning around Events carry a sensitivity marking, but reads are not filtered by it yet: today any credential with `timeline:read` sees every event on your tenant. Treat the change log as shared within your organization and keep anything you would not share out of event payloads. ## Related - [Use the /v1 data endpoints](/knowledge-base/builder-platform/how-to/use-the-v1-endpoints) - [Limits and guardrails](/knowledge-base/builder-platform/reference/limits-and-guardrails) - [Set up a brand](/knowledge-base/plugin/how-to/set-up-a-brand) (how brand context gets authored) --- ### Work with live Amazon operations URL: https://mixshift.ai/knowledge-base/builder-platform/how-to/work-with-live-amazon-operations Markdown: https://mixshift.ai/api/knowledge-base/md/builder-platform/how-to/work-with-live-amazon-operations Type: how-to Product: builder-platform Last modified: 2026-08-28T00:00:00.000Z Call live retail and advertising operations on Amazon through your existing MixShift authorization, including audited writes that preview before they commit. # Work with live Amazon operations Beyond the warehouse, the platform passes through to Amazon itself: live retail lookups, report generation, pricing, and advertising operations, all using the authorization your accounts already have through MixShift. You skip the developer applications entirely, and the token plumbing (minting, refresh, regional routing) is handled server-side. ## Naming the merchant you mean Every call below acts for one merchant. Two identifiers name one, and there is no `merchantId` parameter: - `sellerId`, Amazon's own merchant token (the `A...` / `ENTITY...` string). Enough on its own only when the token maps to a single row; a seller trading in several marketplaces also needs `marketplace`. - `legacySellerId`, the integer MixShift seller id. Identifies exactly one merchant row and takes precedence. Prefer it when you have it. `GET /api/amazon/merchants` returns both for every merchant you can act for, and `GET /api/amazon/spapi/operations` returns the same contract as a `merchantIdentifiers` object next to the operation list, so your code never has to hardcode it. Full detail, including why the same distinction matters when you query the warehouse: [Identify a merchant](/knowledge-base/builder-platform/reference/merchant-identifiers). ## Reading from Amazon - `GET /api/amazon/merchants` lists the merchants your credential can act for. - **Cataloged operations:** `GET /api/amazon/spapi/operations` and `GET /api/amazon/ads/operations` list every callable operation with its parameters; `POST /api/amazon/spapi/call` and `POST /api/amazon/ads/call` invoke one. The catalogs cover retail (catalog, inventory, FBA inbound shipments, orders, finances, fees, pricing, listings) and advertising (campaigns, ad groups, keywords, targets, budgets, recommendations). - **FBA inbound shipments:** read the shipments on their way to Amazon, not just the stock already received. You get shipment status and destination, the items on a shipment across a date range, and the item list for a single shipment id, which is what you need to reconcile what you sent against what Amazon received. - **Reports:** submit a report request, poll it, and download the document through `/api/amazon/reports`. Pricing has its own batch surface under `/api/amazon/pricing`. Report types Amazon has retired fail immediately and name the live replacement to use instead, rather than returning an opaque error from Amazon. ## Creating a Sponsored Brands ad Building a full Sponsored Brands ad from nothing takes an ad group plus one ad-type-specific create, because Amazon exposes a separate endpoint per creative type rather than one shared ad shape. Build in order: create the campaign, create the ad group, then create the ad itself as brand video, video, product collection, or store spotlight, each with its own required creative fields. Two acceptance rules Amazon enforces that a preview cannot check for you: - Every asset id (video, brand logo, custom image) must already exist in the Creative Asset Library. There is no upload operation on this surface, so a new video or logo goes in through the Amazon console first; the Creative Asset Library lookup below gives you the id afterward. - On the ad types that land on your Store, every advertised product must already appear on that Store page, or Amazon rejects the ad. A dry run checks request shape only and never reaches Amazon, so neither rule above is validated until you commit. Amazon validates per ad, so commit one ad first and read its result before sending the rest of a batch. ## Look up Creative Asset Library assets Read the newest page of a profile's Creative Asset Library: asset ids, type, approval status, and file metadata (resolution and duration for video, dimensions and aspect ratio for images). Use it to find the id of something just uploaded in the console, or to check which ad programs an asset already qualifies for before building an ad around it. It returns only the newest page, not the whole library, and the library is scoped per profile, not account-wide: a profile returning no assets does not mean the account has none. ## Advertising writes, with the safety on Writes to your ad accounts are opt-in (the credential needs the write scope) and default to a dry run: 1. Call the write with default settings and you get a **preview**: validation, a snapshot of current state, and the exact change list. Nothing touches Amazon. 2. Send the explicit commit flag to apply. Changes are capped per call, and every commit is audit-logged with the pre-write snapshot. This is the same preview-then-confirm model the [plugin](/knowledge-base/plugin) uses; your own tools get it too. ## Listing content writes Product listing edits (title, highlights, bullets, description, search keywords, ingredients, item form, images) run as a strict two-call contract: 1. **Preview** validates the change against Amazon for up to ten SKUs and returns a change-set id and hash. Nothing is applied. 2. **Commit** takes that id and hash and replays exactly what was previewed. If anything drifted in between, the affected SKU fails closed instead of applying a stale change. Identifiers, brand fields, compliance and safety attributes, variations, price, quantity, fulfillment, and deletes are refused by design. Images apply asynchronously on the Amazon side (Amazon re-hosts them), so verify by checking that the image slot changed and the listing has no new issues, not by comparing URLs. ## Related - [Get access](/knowledge-base/builder-platform/getting-started/get-access) (scopes are granted at mint time) - [Identify a merchant: sellerId and legacySellerId](/knowledge-base/builder-platform/reference/merchant-identifiers) - [Limits and guardrails](/knowledge-base/builder-platform/reference/limits-and-guardrails) - Operation-level reference: [mcp.mixshift.io/developers](https://mcp.mixshift.io/developers) --- ### Identify a merchant, sellerId and legacySellerId URL: https://mixshift.ai/knowledge-base/builder-platform/reference/merchant-identifiers Markdown: https://mixshift.ai/api/knowledge-base/md/builder-platform/reference/merchant-identifiers Type: reference Product: builder-platform Last modified: 2026-08-25T00:00:00.000Z The two identifiers that name an Amazon merchant on the Builder Platform: sellerId (Amazon's merchant token) and legacySellerId (the integer seller id). There is no merchantId. The same distinction decides whether a warehouse query is fast or slow. # Identify a merchant: `sellerId` and `legacySellerId` Almost everything on the Builder Platform acts for one Amazon merchant, and there are exactly two ways to name one. Picking the wrong one is an easy first-call mistake, and it is worth thirty seconds up front because it costs you twice: once on the API, where the call fails or resolves to the wrong merchant, and once in the warehouse, where the same mistake turns a fast query into a slow one without changing the answer. ## The two identifiers | Identifier | Type | What it is | | --- | --- | --- | | `sellerId` | string | Amazon's own merchant token, the `A...` or `ENTITY...` string. Stored in the warehouse as `AmazonSellerID`. | | `legacySellerId` | integer | The MixShift seller id. The durable unique key for exactly one merchant row, meaning one seller in one marketplace. Stored in the warehouse as `SellerID`. | Both come back from `GET /api/amazon/merchants`, one row per account and marketplace. **There is no `merchantId` parameter.** It has never existed on these endpoints. If you passed one, it was ignored, and the call either resolved to some other merchant or failed to resolve at all. ## Which one to pass Prefer `legacySellerId` whenever you have it. It identifies exactly one merchant row and takes precedence over everything else. `sellerId` is fine on its own **when the token maps to a single row**. A seller who trades in more than one marketplace has several rows behind one token, so `sellerId` alone is ambiguous there. Add `marketplace` (`US`, `CA`, `MX`, and so on) to narrow it, or use `legacySellerId` and skip the question. ```bash # Unambiguous: one merchant row. curl -X POST https://mcp.mixshift.io/api/amazon/spapi/call \ -H "Authorization: Bearer $TOKEN" \ -H "Content-Type: application/json" \ -d '{"operation": "fulfillment_inbound.get_shipments", "legacySellerId": 12345, "query": {"QueryType": "SHIPMENT", "ShipmentStatusList": "RECEIVING,CLOSED"}}' ``` If a merchant cannot be resolved, the error tells you which identifiers the endpoint accepts and lists the candidate rows when the problem is ambiguity rather than a bad id. ## The same distinction in the warehouse Most seller-scoped warehouse tables carry **both** columns: an integer `SellerID` and a varchar `AmazonSellerID`. They identify the same seller and return the same rows, but they are not interchangeable for speed, because the indexes are generally built on the integer. **Filter on the integer `SellerID`.** Filtering on `AmazonSellerID` often means the composite index that pairs the seller with a date is not used, which leaves a date range with nothing to seek on and scans every row for that seller. `spapi_settlement` is the clearest case. The two indexes that matter here are: - `MIX_MAX_Record_IDX` on (`SellerID`, `Posted-date-time`) - `Delete_Settlement_IDX` on (`AmazonSellerID`, `Settlement-id`) So a settlement query filtered by date only has an index to work with if it filters on the integer `SellerID`. Measured on one account, on one tenant database, with the identical one-month count returning exactly the same rows either way: ```sql -- 13.4 seconds. The date predicate has no index to use. SELECT COUNT(*) FROM spapi_settlement WHERE AmazonSellerID = 'A1...' AND `Posted-date-time` >= '2026-07-01' AND `Posted-date-time` < '2026-08-01'; -- 0.37 seconds. Covered by MIX_MAX_Record_IDX. SELECT COUNT(*) FROM spapi_settlement WHERE SellerID = 12345 AND `Posted-date-time` >= '2026-07-01' AND `Posted-date-time` < '2026-08-01'; ``` Same answer, roughly 36 times the wait. That is what makes this worth knowing: nothing looks broken, so the usual next question is whether the table is indexed. It is. Just not for the column you filtered on. Index names and layouts live in each tenant's own database, so treat the shape as the lesson rather than the specific names. Reach for `AmazonSellerID` when you are looking up a specific `Settlement-id`, which is what its index is for. Otherwise translate the token to the integer once at the start of your job: ```sql SELECT ID FROM seller WHERE AmazonSellerID = 'A1...'; ``` `GET /api/amazon/merchants` gives you the same mapping without a query. This is a strong default rather than a universal law. If a query on a large table is slower than you expect, run `EXPLAIN` on it and check whether the chosen key actually covers your date predicate before concluding the table is the problem. ## Related - [Work with live Amazon operations](/knowledge-base/builder-platform/how-to/work-with-live-amazon-operations) - [Query the warehouse](/knowledge-base/builder-platform/how-to/query-the-warehouse) - [Limits and guardrails](/knowledge-base/builder-platform/reference/limits-and-guardrails) - Operation-level reference: [mcp.mixshift.io/developers](https://mcp.mixshift.io/developers) --- ### Limits and guardrails URL: https://mixshift.ai/knowledge-base/builder-platform/reference/limits-and-guardrails Markdown: https://mixshift.ai/api/knowledge-base/md/builder-platform/reference/limits-and-guardrails Type: reference Product: builder-platform Last modified: 2026-08-28T00:00:00.000Z The published limits (rate, rows, size, time), the permission scopes, and the guardrails that make the platform safe to hand to your own tools. # Limits and guardrails The platform is designed so that the safe path is the default path. This page is the honest fine print. ## Published limits - **240 requests per minute per credential.** Exceeding it returns a 429 with a Retry-After header. - **SQL results:** up to 50,000 rows or 10 MB per response, 120 second query ceiling (60 by default). Paginate or chunk by date beyond that. - **Access tokens are short-lived** and refresh through the standard token endpoint. Service-credential tokens last **12 hours**, so an unattended job that mints one at startup does not need to refresh mid-run. Partner API sessions run the same 12-hour ceiling, matching service credentials. Revoking a credential still stops any new token from being issued; an outstanding token simply runs out within that same ceiling. Amazon-side throttling surfaces as its own error kind so you can back off correctly. These are generous for interactive use; if you hit them in an automation, batch your requests. ## Scoped credentials Every credential carries explicit permission scopes: account, retail, advertising, brand analytics, context, and timeline reads are separate grants, and the three write permissions (timeline, advertising, listings) are opt-in at mint time. Raw SQL against the warehouse is its own scope, `sql:query`, separate from the domain reads above; it is bundled by default when a credential is minted with no explicit scope list, but a credential minted with an explicit list needs to request it by name. Every call re-checks its scope, so a read-only credential cannot write no matter what the calling code (or the AI model behind it) asks for. Credentials can be rotated or revoked at any time. ## Your data only A credential resolves server-side to your own tenant database and your own authorized Amazon accounts. The isolation is structural, not a filter: there is no parameter that reaches anyone else's data. ## Writes preview first Advertising and listing writes validate, snapshot, and preview before anything is applied, and commits are audit-logged with the pre-write state. Listing commits replay exactly what was previewed and fail closed if anything drifted. See [Work with live Amazon operations](/knowledge-base/builder-platform/how-to/work-with-live-amazon-operations). ## Predictable errors Failures return a consistent envelope with a stable machine-readable kind (insufficient scope, throttled, merchant not found, and so on) plus a human-friendly message. Branch on the kind; show the message. One part of that contract is worth building around deliberately: a failed write does not always mean nothing happened. A few failure kinds are ambiguous by nature, because none of them proves the change was rejected before it was applied: a server error returned from Amazon, a throttle (which can arrive after acceptance), and no response at all. On those, verify the current state before you act, and never blindly re-send the change, or you can apply it twice. Every other kind is a definite rejection, and there it is safe to assume nothing was applied. The operation-level rules are on [mcp.mixshift.io/developers](https://mcp.mixshift.io/developers). ## Related - [Get access](/knowledge-base/builder-platform/getting-started/get-access) - [Query the warehouse](/knowledge-base/builder-platform/how-to/query-the-warehouse) - [Builder Platform FAQ](/knowledge-base/builder-platform/faq/faq) --- ### When a call fails URL: https://mixshift.ai/knowledge-base/builder-platform/reference/when-a-call-fails Markdown: https://mixshift.ai/api/knowledge-base/md/builder-platform/reference/when-a-call-fails Type: reference Product: builder-platform Last modified: 2026-08-26T00:00:00.000Z How to read a failure from the Builder Platform: whose problem it is, whether retrying will ever work, what Amazon actually said, and how to report a failure our own documentation caused. # When a call fails Every failure comes back in the same envelope: ```json { "ok": false, "kind": "bad_request", "friendly": "...", "message": "...", "amazon_error_code": "InvalidInput", "status": 400 } ``` Branch on `kind`, not on the HTTP status. `friendly` is safe to show a person as-is. `message` is the specific reason, and it is the field to read when you are trying to fix something. ## The only question that matters first: whose problem is it? Retrying the wrong kind of failure is the most common way to lose an afternoon here, so the answer is meant to be obvious from the response alone. | You get | Whose problem | Retry? | | --- | --- | --- | | `400` `bad_request` | **Yours.** Amazon rejected the request itself. | **No.** It is deterministic and will fail identically forever. Change the request. | | `403` `restricted_report` | An access or role gap. | No. Re-authorize the merchant, or use a different operation. | | `409` `reauth_required` | The merchant's Amazon connection lapsed. | Re-connect the account first, then yes. | | `404` `merchant_not_found` | The merchant selector matched nothing. | No. See [Identify a merchant](/knowledge-base/builder-platform/reference/merchant-identifiers). | | `429` `throttled` | Rate limiting, ours or Amazon's. | Yes, after backing off. Honour `Retry-After`. | | `502` `upstream_unavailable` or `host_unreachable` | Amazon's, or the network. | Yes, later. **For writes, check whether it applied first.** | | `500` `unknown` | Ours, and now rare. | Once. If it persists, tell us. | The write caveat on `502` is worth reading twice: a server error does not prove Amazon rejected your change before acting on it. Re-read the entity before you re-send, or you can apply the same change twice. ## `bad_request`: what Amazon actually said This is the one to know, because it used to be the confusing one. A request Amazon refused because of its own parameters used to come back as a `500`, which reads as an outage and invites a retry that could never work. It now comes back as a `400` and carries Amazon's own words: - `amazon_error_code` is Amazon's code, such as `InvalidInput` or `InvalidParameterValue`. - `status` is Amazon's own HTTP status, separate from ours. - `responsePayload` is Amazon's full response, when we captured it. In the plugin, the same information prints to the terminal, and `--json` carries `amazon_error_code`, `amazon_status` and `amazon_response`. Scripts get exit code `12`, which is distinct from the generic failure code precisely so a script can stop instead of retrying. Most of the time this means a parameter is missing, misspelled, or in the wrong case. The operation catalog (`GET /api/amazon/spapi/operations`) documents the required parameters and the casing rules for each call, and it is the first thing to check. ## Reporting a failure, and why it is worth your thirty seconds Sometimes the request was wrong **because our documentation said to make it that way.** The operation catalog is maintained by hand. Amazon occasionally enforces a parameter it does not document, or requires two parameters only in combination, and until someone hits it our notes look complete. When our notes are wrong, every caller walks into the same wall. Here is the asymmetry that makes your report valuable: **we can see that a call failed and which operation it was. We cannot see your parameters.** You are holding the half we are missing. So please send three things: 1. the **operation id**, for example `fulfillment_inbound.get_shipments` 2. the **`amazon_error_code`** and the `message` from the envelope 3. the **parameters you sent**, minus anything you consider sensitive Where to send it: - **Plugin users:** run `mixshift feedback`. It attaches the context for you. - **Everyone else:** [support@mixshift.io](mailto:support@mixshift.io). We record the Amazon error code against the operation, so the same code repeating on the same call surfaces as a pattern on our side. A confirmed gap normally turns into a catalog note that the next caller never trips over. That is the whole loop, and it starts with someone bothering to tell us. ## Related - [Identify a merchant: sellerId and legacySellerId](/knowledge-base/builder-platform/reference/merchant-identifiers) - [Limits and guardrails](/knowledge-base/builder-platform/reference/limits-and-guardrails) - [Work with live Amazon operations](/knowledge-base/builder-platform/how-to/work-with-live-amazon-operations) - Operation-level reference: [mcp.mixshift.io/developers](https://mcp.mixshift.io/developers) --- ### Builder Platform FAQ URL: https://mixshift.ai/knowledge-base/builder-platform/faq/faq Markdown: https://mixshift.ai/api/knowledge-base/md/builder-platform/faq/faq Type: faq Product: builder-platform Last modified: 2026-07-31T00:00:00.000Z Common questions about building on MixShift: cost, reachable data, writes, Amazon registration, MySQL migration, MCP, and multi-user apps. # Builder Platform FAQ ## What does it cost? Nothing extra. Builder Platform access is included with your MixShift subscription. ## What data can I reach? Your whole MixShift warehouse over SQL (advertising, DSP, Seller and Vendor Central revenue and orders, inventory, catalog), the pre-shaped `/v1` endpoints (accounts, advertising performance, pacing, search terms, ASIN performance, retail sales and traffic, Search Query Performance, brand context and change log), and live operations against Amazon for the accounts you are authorized on. ## Can my app read the brand context and change log? Yes, and this is the part most people do not realize is there. Your brand context documents and the brand change log (including stakes: the launches, promotions, stockouts, and price tests that explain a move) are on `/v1` and MCP with the same credential. Reads are included in the default read scopes; appending to the change log is an opt-in write scope. See [Work with brand memory](/knowledge-base/builder-platform/how-to/work-with-brand-memory). ## Can I get MixShift Intelligence through the API? MixShift Intelligence (the insight engine behind the bridge reads, estimated lost sales, and the monthly read) is in closed beta and switched on per account. It is not part of the documented `/v1` partner contract today; enabled accounts reach it through MCP tools and the plugin command line. If you want your account considered for early access, write to support@mixshift.io or mention it to your MixShift contact. ## Can my app write to my ad accounts? Yes, if the credential was minted with the write permission. Writes default to a dry run that previews the exact change, and commits are capped and audit-logged with a pre-write snapshot. Listing content writes use a stricter two-call preview and commit contract. Read-only credentials cannot write at all. ## Do I need to register with Amazon as a developer? No. That is most of the point. Your accounts are already authorized through MixShift, so you skip the developer applications, and the token plumbing is handled server-side. ## I already query the warehouse over MySQL. Why switch? The HTTPS endpoint runs the same SQL without database passwords, driver quirks, or IP allowlists, and it works from anywhere (serverless included). Direct MySQL access is the legacy path; new builds should start on `/api/query`. ## Do I have to use the MCP? No. MCP is one of three doors. If you are writing code, the REST endpoints are probably what you want; MCP exists so AI tools can connect without you writing anything. ## Which AI tools can connect over MCP? Claude, ChatGPT, Cursor, Codex, and any MCP-compatible client, all against the same server with one credential. See [Connect AI tools over MCP](/knowledge-base/builder-platform/how-to/connect-ai-tools-over-mcp). ## Can a multi-user application sit on one credential? Yes for internal tools, where the credential represents your organization. Partner software with per-customer sign-in should use the registered-app path instead, so each customer consents and gets their own tokens. See [Get access](/knowledge-base/builder-platform/getting-started/get-access). ## How do I get access? [Get in touch](/contact?interest=builder-platform) and we will set you up, or your MixShift admin can mint a service credential directly. Details in [Get access](/knowledge-base/builder-platform/getting-started/get-access). ## Related - [Limits and guardrails](/knowledge-base/builder-platform/reference/limits-and-guardrails) - [Query the warehouse](/knowledge-base/builder-platform/how-to/query-the-warehouse) --- ### Data Hub URL: https://mixshift.ai/knowledge-base/data-hub Markdown: https://mixshift.ai/api/knowledge-base/md/data-hub Type: index Product: data-hub Last modified: 2026-07-31T00:00:00.000Z How MixShift connects to your Amazon accounts and keeps the data fresh, and how to check the health of a connection. # Data Hub How MixShift connects to your Amazon data and keeps it fresh. Data Hub is in beta. ## Concept - [How MixShift keeps your Amazon data fresh](/knowledge-base/data-hub/concept/how-mixshift-keeps-your-data-fresh) ## How to - [Check your data health](/knowledge-base/data-hub/how-to/check-your-data-health) --- ### Check your data health URL: https://mixshift.ai/knowledge-base/data-hub/how-to/check-your-data-health Markdown: https://mixshift.ai/api/knowledge-base/md/data-hub/how-to/check-your-data-health Type: how-to Product: data-hub Last modified: 2026-07-21T00:00:00.000Z Use the Data Hub page to see connection health, per-merchant sync status, and data freshness at a glance. # Check your data health If a report looks off, the first question is always the same: is the data behind it current? The Data Hub page answers that in one glance, without a support ticket. Data Hub is in beta and is visible to organization Owners and Admins. ## Open Data Hub Go to the Data Hub page in your account area. If you belong to more than one organization, you see the one you are currently working in. ## What each section tells you - **Connector health.** The state of your Selling Partner and Advertising connections. If one has expired, this is where it shows first, with the option to reconnect. - **Merchant sync status.** One row per connected merchant: when it last synced and whether the latest sync succeeded. - **Task health.** Whether the background sync jobs are running normally. - **Table freshness.** How current each data table is, so you can tell "the sync ran" apart from "the data is up to date." ## Reading it when something is wrong 1. A report looks empty or stale? Check connector health first. An expired connection pauses syncing for that merchant. 2. Connection healthy but a merchant looks behind? Check its row in merchant sync status; the latest run may have failed and will retry. 3. Everything green but numbers seem old? Check table freshness. Data updates once per day, so "through yesterday" is normal. ## Notes - Data Hub shows the state of your own organization only. - If a connection repeatedly fails after reconnecting, contact support; that is not something you can fix from your side. ## Related - [How MixShift keeps your data fresh](/knowledge-base/data-hub/concept/how-mixshift-keeps-your-data-fresh) - [Report Center troubleshooting](/knowledge-base/report-center/troubleshooting/troubleshooting) --- ### How MixShift keeps your Amazon data fresh URL: https://mixshift.ai/knowledge-base/data-hub/concept/how-mixshift-keeps-your-data-fresh Markdown: https://mixshift.ai/api/knowledge-base/md/data-hub/concept/how-mixshift-keeps-your-data-fresh Type: concept Product: data-hub Last modified: 2026-07-21T00:00:00.000Z What MixShift pulls from Amazon, how often, and what happens when a connection needs attention. # How MixShift keeps your Amazon data fresh Every report you see in MixShift is built from data we pull from Amazon on your behalf. Here is how that works, in plain terms. ## What we pull - **Retail data** comes from your Selling Partner connection: orders, sales, and financial events for each connected merchant and marketplace. - **Advertising data** comes from your Amazon Advertising connection (optional, recommended): campaigns, spend, and performance metrics. You authorize both connections once, through Amazon, during setup. You can add or reconnect them any time from Settings. ## How often Your data syncs once per day, in the background. Reports reflect your most recent nightly sync, so the numbers you see in the morning cover through yesterday. Real-time data is not available today. ## When something needs attention Amazon connections can expire; that is how Amazon works, not a bug. When a connection expires, syncing pauses for that merchant until you reconnect. You will see the connection marked as needing attention, and reconnecting from Settings resumes the sync. Nothing is lost while a connection is paused; the next sync catches up. ## Seeing it for yourself If you are an organization Owner or Admin, the Data Hub page shows the health of every connection, per-merchant sync status, and how fresh each data table is. See [Check your data health](/knowledge-base/data-hub/how-to/check-your-data-health). ## Related - [Check your data health](/knowledge-base/data-hub/how-to/check-your-data-health) - [Report Center troubleshooting](/knowledge-base/report-center/troubleshooting/troubleshooting) --- ### DSP Managed Service URL: https://mixshift.ai/knowledge-base/dsp-managed-service Markdown: https://mixshift.ai/api/knowledge-base/md/dsp-managed-service Type: index Product: dsp-managed-service Last modified: 2026-07-21T00:00:00.000Z Amazon DSP advertising run for you by the MixShift team, with your reports one question away. # DSP Managed Service Amazon DSP (the display advertising platform from Amazon) run for you by the MixShift team: strategy, setup, and ongoing management, with reporting you can pull yourself any time. ## How to - [Pull your DSP reports in Claude](/knowledge-base/dsp-managed-service/how-to/pull-dsp-reports-in-claude) Interested in the managed service itself? [Get in touch](/knowledge-base/shared/how-to/contact-support) and we will walk you through how it works. More articles are on the way. --- ### Pull your DSP reports in Claude URL: https://mixshift.ai/knowledge-base/dsp-managed-service/how-to/pull-dsp-reports-in-claude Markdown: https://mixshift.ai/api/knowledge-base/md/dsp-managed-service/how-to/pull-dsp-reports-in-claude Type: how-to Product: dsp-managed-service Last modified: 2026-07-21T00:00:00.000Z Ask for Amazon DSP campaign, audience, and inventory reports in plain English through the MixShift plugin. # Pull your DSP reports in Claude If MixShift runs Amazon DSP (the display advertising platform from Amazon) for you, or your account has its own DSP seat, you can pull DSP reports by asking for them in a Claude chat. No logging into the DSP console, no report builder. ## Before you start - An Amazon advertising account with DSP access. - The MixShift plugin installed and signed in. See [Install the plugin](/knowledge-base/plugin/getting-started/install-the-plugin). ## What you can ask for - **Campaign reports.** "Pull a DSP campaign report for last week." - **Audience reports.** "What did our DSP audience segments do this month?" - **Inventory reports.** "Show DSP inventory performance by supply source." Claude requests the report from Amazon, waits for Amazon to generate it (these are built on the Amazon side, so allow a little time), and returns the results in the chat with a downloadable copy. ## Notes - Read-only. Pulling reports changes nothing in your campaigns. - If Claude reports that no DSP advertiser is found, your DSP may run under a separate seat. Mention that to your MixShift contact; it is a known setup, not a dead end. - DSP reporting is in beta in the plugin. ## Related - [Install the plugin](/knowledge-base/plugin/getting-started/install-the-plugin) - [Contact MixShift support](/knowledge-base/shared/how-to/contact-support) --- ### Forecasting & HCAM URL: https://mixshift.ai/knowledge-base/forecasting-hcam Markdown: https://mixshift.ai/api/knowledge-base/md/forecasting-hcam Type: index Product: forecasting-hcam Last modified: 2026-07-21T00:00:00.000Z How MixShift forecasts your business and explains why your numbers moved. In development. # Forecasting & HCAM MixShift forecasts where your business is heading and explains why your numbers moved, using our attribution approach (HCAM). We are writing a plain-language explainer of how it works, plus a glossary of the metrics you will see. Articles will land here as they are ready. In the meantime, ask Shifty or [get in touch](/knowledge-base/shared/how-to/contact-support). --- ### MixShift AI Plugin URL: https://mixshift.ai/knowledge-base/plugin Markdown: https://mixshift.ai/api/knowledge-base/md/plugin Type: index Product: plugin Last modified: 2026-07-31T00:00:00.000Z MixShift skills, warehouse data, and live Amazon operations inside Claude. Install, sign in, set up a brand, and put the skill catalog to work. # MixShift AI Plugin The MixShift AI plugin puts your MixShift data and a library of purpose-built skills inside Claude. Ask questions about your brands in plain English, pull reports from Amazon on demand, run daily health checks and search-term reviews, and make audited changes to your ad accounts, all from a chat. The plugin is in beta and included with your MixShift subscription. It runs where Claude runs for you: the Cowork desktop app, Claude Code, or rolled out to your whole company by an admin. Other AI tools (ChatGPT, Cursor, Codex) can reach the same data through the [Builder Platform](/knowledge-base/builder-platform). ## Getting started - [Install the plugin](/knowledge-base/plugin/getting-started/install-the-plugin) - [Sign in to MixShift](/knowledge-base/plugin/getting-started/sign-in) ## How to - [Set up a brand](/knowledge-base/plugin/how-to/set-up-a-brand) - [Run a skill](/knowledge-base/plugin/how-to/run-a-skill) - [Install for your whole team](/knowledge-base/plugin/how-to/install-for-your-whole-team) - [Keep the plugin updated](/knowledge-base/plugin/how-to/keep-the-plugin-updated) - [Run unattended and scheduled tasks](/knowledge-base/plugin/how-to/run-unattended-and-scheduled-tasks) - [Run AMC queries from Claude](/knowledge-base/plugin/how-to/run-amc-queries-from-claude) ## Reference - [Skill catalog](/knowledge-base/plugin/reference/skill-catalog) - [What the plugin can do](/knowledge-base/plugin/reference/what-the-plugin-can-do) ## FAQ - [Plugin FAQ](/knowledge-base/plugin/faq/faq) ## Troubleshooting - [Troubleshooting](/knowledge-base/plugin/troubleshooting/troubleshooting) --- ### Install the MixShift plugin URL: https://mixshift.ai/knowledge-base/plugin/getting-started/install-the-plugin Markdown: https://mixshift.ai/api/knowledge-base/md/plugin/getting-started/install-the-plugin Type: getting-started Product: plugin Last modified: 2026-07-31T00:00:00.000Z Install the MixShift AI plugin in Cowork or Claude Code, or roll it out company-wide. Pick your path and be running in about five minutes. # Install the MixShift plugin The plugin installs from the MixShift catalog on GitHub. Pick the path that matches how you use Claude; each takes a few minutes, and every path ends the same way: you sign in once and start asking questions about your data. ## Which path is right for me? - **You use the Cowork desktop app on your own.** Follow the Cowork steps below. - **Your team uses Cowork and you want everyone to have it.** An org admin publishes it once. See [Install for your whole team](/knowledge-base/plugin/how-to/install-for-your-whole-team). - **Your company manages Claude centrally (Team or Enterprise plan).** Your Claude admin adds it through the admin console. See the admin console note below. - **You use Claude Code** (the terminal or desktop app). Two commands, below. - **You want the MixShift CLI without Claude.** See the CLI note at the bottom. - **You use claude.ai in a browser, or the Claude Desktop app.** The plugin does not install there. Connect MixShift as an MCP connector instead and you get the same data in chat, without the packaged skills: see [Connect AI tools over MCP](/knowledge-base/builder-platform/how-to/connect-ai-tools-over-mcp). We recommend the **stable channel** for everyone. It is the tested release cut; installs that skip it track the newest changes before they are announced. ## Before you start - An active MixShift account. You sign in with your MixShift work email. - Node.js version 20 or newer on your computer. To check, open a terminal and run `node --version`. ## Install in Cowork (personal) 1. In Cowork, click **Customize** in the left sidebar. 2. Click the **+** button and choose **Add marketplace from GitHub**. 3. Paste `https://github.com/miXshift/mx-claude-plugin/tree/stable` and confirm. This registers the MixShift catalog on your seat only. 4. Open the **Directory** view (Skills / Connectors / Plugins under Customize). 5. Find **mixshift-ai** in the plugin list and click **Install**. 6. In the same view, turn on **Sync automatically** for the MixShift marketplace so you keep getting updates. The plugin appears under **Personal plugins** in your sidebar. ## Install in Claude Code Run these two commands inside Claude Code: 1. `/plugin marketplace add https://github.com/miXshift/mx-claude-plugin.git#stable` 2. `/plugin install mixshift-ai@mixshift` Use the full address with `#stable` as written; short forms pin you to the wrong channel. ## Company-wide installs - **Cowork organization:** an org admin publishes the plugin once from Organization settings and every seat gets it. See [Install for your whole team](/knowledge-base/plugin/how-to/install-for-your-whole-team). - **Claude admin console (Team and Enterprise plans):** your Claude admin uploads the MixShift marketplace zip from the GitHub Releases page, or points a private mirror repository at our stable channel so updates flow automatically. Ask us if your admin wants a hand; this path takes minutes. ## CLI without Claude Builders and scripts can use the bundled `mixshift` command line directly: clone the repository, then build the harness (`npm install` and `npm run build` inside the harness folder) and put `harness/bin/mixshift` on your PATH. This needs `git` and Node 20 or newer. Most people do not need this path; the plugin installs the CLI for you. ## Next Sign in and try your first question: [Sign in to MixShift](/knowledge-base/plugin/getting-started/sign-in). --- ### Sign in to MixShift URL: https://mixshift.ai/knowledge-base/plugin/getting-started/sign-in Markdown: https://mixshift.ai/api/knowledge-base/md/plugin/getting-started/sign-in Type: getting-started Product: plugin Last modified: 2026-07-31T00:00:00.000Z Sign in from any Claude chat in about 30 seconds. Your password stays on the sign-in page; the plugin only ever holds short-lived tokens. # Sign in to MixShift Once the plugin is installed, you sign in once and the plugin keeps your session fresh from then on. The whole flow takes about 30 seconds. ## Sign in from a chat 1. Open any chat and say **"sign in to mixshift"** (or just "welcome" if you are brand new). 2. Claude asks for your work email, then opens the MixShift sign-in page in a browser tab. 3. Sign in with your MixShift email and password. Your credentials stay on the sign-in page; the plugin never sees them. 4. Return to the chat and say "done". Claude confirms the sign-in and shows the brands you have access to. ## Try it - "What brands do I have access to?" - "Show me last week's ad spend for [your brand]." - "Explore my data." If Claude returns your brands and data, you are set. ## Good to know - Sign-in uses short-lived tokens that refresh automatically. There are no database passwords to manage and no IP allowlists to set up. - To switch accounts, just sign in again with the other email. - Everyone at your MixShift org sees the same brands once signed in. Per-user data scoping is not available today. - In a terminal, `mixshift auth login` does the same thing for CLI users. ## Signing in from other tools Cursor, Codex, ChatGPT, and claude.ai connectors reach MixShift through the MCP server rather than the plugin. See [Connect AI tools over MCP](/knowledge-base/builder-platform/how-to/connect-ai-tools-over-mcp). ## Unattended and scheduled work Anything that runs with nobody at the keyboard (scheduled tasks, automations, CI) uses a service credential instead of a personal sign-in. See [Run unattended and scheduled tasks](/knowledge-base/plugin/how-to/run-unattended-and-scheduled-tasks). ## Related - [Install the plugin](/knowledge-base/plugin/getting-started/install-the-plugin) - [Set up a brand](/knowledge-base/plugin/how-to/set-up-a-brand) --- ### Install the plugin for your whole team URL: https://mixshift.ai/knowledge-base/plugin/how-to/install-for-your-whole-team Markdown: https://mixshift.ai/api/knowledge-base/md/plugin/how-to/install-for-your-whole-team Type: how-to Product: plugin Last modified: 2026-07-31T00:00:00.000Z An admin publishes the plugin once for the whole organization, then each teammate signs in with their own MixShift account. # Install the plugin for your whole team You can make the MixShift plugin available to every seat at once instead of each person installing it themselves. There are two parts: an admin publishes the plugin once, then each person signs in with their own MixShift account. ## Before you start - You need to be an owner or admin of your Claude organization. - Your team members each need an active MixShift account to sign in with. ## Option A: Cowork organization (admin, once) 1. In Cowork, go to **Organization settings**, then **Plugins**. 2. Click **Add plugin** and select **GitHub source**. 3. Enter `miXshift/mx-claude-plugin` and save. 4. Optional: mark the plugin as **required** so it installs automatically for every seat. Otherwise each person clicks Install when they want it. The plugin now appears for everyone in your organization's plugin directory. ## Option B: Claude admin console (Team and Enterprise plans) If your company manages Claude through the central admin console: 1. Download the MixShift marketplace zip from the plugin repository's GitHub Releases page, and upload it in the console's plugin section. Re-upload when you want to take an update, **or** 2. Set up a private mirror repository that auto-syncs our stable channel, so updates flow to your org without re-uploads. Ask us if your admin wants a hand; either path takes minutes. ## Part 2: Each person signs in (about 30 seconds) Everyone on the team, individually: 1. Opens any chat and says **"sign in to mixshift"**. 2. Claude opens the MixShift sign-in page in a browser tab. 3. They sign in with their own MixShift email and password, return to the chat, and say "done". There are no shared passwords to distribute and no credential files to manage. Each person's sign-in is their own. ## Verify the rollout Ask a teammate who has signed in to try: "What brands do I have access to?" If Claude lists your brands, the rollout works. ## Notes - Everyone at your MixShift org sees the same brands once signed in. Per-user data scoping is not available today. - Updates ship through the same channel you installed from. See [Keep the plugin updated](/knowledge-base/plugin/how-to/keep-the-plugin-updated). - Brands set up with [brand setup](/knowledge-base/plugin/how-to/set-up-a-brand) publish to your organization store, so one teammate's setup benefits everyone. ## Related - [Install the plugin (personal paths)](/knowledge-base/plugin/getting-started/install-the-plugin) - [Sign in to MixShift](/knowledge-base/plugin/getting-started/sign-in) --- ### Keep the plugin updated URL: https://mixshift.ai/knowledge-base/plugin/how-to/keep-the-plugin-updated Markdown: https://mixshift.ai/api/knowledge-base/md/plugin/how-to/keep-the-plugin-updated Type: how-to Product: plugin Last modified: 2026-07-31T00:00:00.000Z Ask "update the plugin" and Claude walks you through it, or update manually per surface. What's new, in chat, whenever you ask. # Keep the plugin updated The plugin ships improvements weekly. Two chat phrases cover most of what you need: - **"What's new in mixshift?"** shows the changes since your installed version. - **"Update the mixshift plugin"** walks you through the update for your surface and then tells you if anything needs a follow-up step. It compares the build this session is running against the build installed on your machine, so "the update landed, you just need to relaunch" is told apart from "the update did not land." ## Manual update, by surface - **Claude Code:** run `/plugin marketplace update mixshift`, then `/plugin update mixshift-ai`. - **Cowork (personal install):** open the **Directory** view, find the MixShift marketplace, open the three-dot menu, and choose **Check for updates**. If you turned on **Sync automatically** at install time, updates arrive on their own. - **Cowork (organization install):** your org admin republishes; each seat picks it up. - **Claude admin console:** re-upload the release zip, or nothing at all if your admin set up the auto-syncing private mirror. ## The one rule everyone forgets **A running session keeps the version it started with.** Starting a new chat or session is not enough on its own. Fully quit the application, then relaunch it, so the updated build is the one that loads. If the old version still shows after a full quit and relaunch, an earlier copy of the application is usually still running in the background and still serving the old build; close that one too. If skills behave oddly right after updating, this is almost always why. ## Good to know - The stable channel is the tested weekly cut; release notes live on [the releases page](/releases). - In Cowork, the version label in the plugin panel sometimes lags after an update. If it looks stale, remove and re-add the marketplace entry; your settings and sign-in survive. ## Related - [Install the plugin](/knowledge-base/plugin/getting-started/install-the-plugin) - [Troubleshooting](/knowledge-base/plugin/troubleshooting/troubleshooting) --- ### Run a skill URL: https://mixshift.ai/knowledge-base/plugin/how-to/run-a-skill Markdown: https://mixshift.ai/api/knowledge-base/md/plugin/how-to/run-a-skill Type: how-to Product: plugin Last modified: 2026-07-31T00:00:00.000Z Two ways to run any MixShift skill, what to expect while it works, and how changes are always previewed before anything reaches Amazon. # Run a skill Every MixShift skill runs the same two ways. Use whichever feels natural. ## Ask in plain English Describe what you want and name the brand where it matters: - "Run the daily health check for [brand]." - "Pull last month's Sales and Traffic report for [brand]." - "Which search terms should we negate for [brand]?" Claude picks the right skill and confirms before doing anything consequential. ## Or invoke it directly Each skill has a slash command in the form `/mixshift-ai:mx-[skill]`. Brand skills take the brand as an argument, for example: - `/mixshift-ai:mx-daily-health-check [brand]` - `/mixshift-ai:mx-data-explore` The [skill catalog](/knowledge-base/plugin/reference/skill-catalog) lists every command. ## What to expect - **Changes preview first.** Any operation that would change something in your ad account shows you the exact change set and waits for your confirmation. If you decline, nothing is sent to Amazon. - **Big results become files.** Large query results and report downloads save as CSV files with a short preview in chat, so your conversation stays readable and the full data is on disk. - **Slow pulls are normal for some sources.** Amazon generates some reports asynchronously (AMC especially); the skill submits, keeps checking, and tells you when results are in. - **Some skills want brand context.** The analytical tier reads your brand profile; run [brand setup](/knowledge-base/plugin/how-to/set-up-a-brand) once per brand first. ## Finding the right skill Say **"mixshift help"** or run `/mixshift-ai:mx-help` for a guided capability map, grouped by what you are trying to get done. The same map lives in the [skill catalog](/knowledge-base/plugin/reference/skill-catalog). ## Related - [Skill catalog](/knowledge-base/plugin/reference/skill-catalog) - [Set up a brand](/knowledge-base/plugin/how-to/set-up-a-brand) - [What the plugin can do](/knowledge-base/plugin/reference/what-the-plugin-can-do) --- ### Run AMC queries from Claude URL: https://mixshift.ai/knowledge-base/plugin/how-to/run-amc-queries-from-claude Markdown: https://mixshift.ai/api/knowledge-base/md/plugin/how-to/run-amc-queries-from-claude Type: how-to Product: plugin Last modified: 2026-07-31T00:00:00.000Z Ask Amazon Marketing Cloud questions in plain English and get results back as CSV files, through the MixShift plugin. # Run AMC queries from Claude Amazon Marketing Cloud (AMC) is a clean room from Amazon: a place where you can analyze detailed advertising signals that never appear in normal reports, with results returned only as aggregates, so no individual shopper is ever identifiable. Working with AMC normally means writing SQL against its tables yourself. With the MixShift plugin, you ask in plain English and Claude handles the SQL. ## Before you start - Your advertising account needs AMC access (an AMC instance). If you are not sure you have one, ask; the plugin can check. - The MixShift plugin installed and signed in. See [Install the plugin](/knowledge-base/plugin/getting-started/install-the-plugin). ## How it works 1. Ask a question in a Claude chat, for example "What AMC instances do I have?" or "Using AMC, how many of last month's purchasers saw both our display ads and our sponsored ads?" 2. Claude finds your AMC instance, looks at the data sources available in it, and writes the query. 3. The query runs on the Amazon side. AMC queries are not instant; expect minutes, not seconds. Claude checks progress and tells you when it is done. 4. Results come back as a downloadable CSV you can open in a spreadsheet or hand to another tool. ## Good first questions - "List my AMC instances and what data sources they have." - "How much did display and sponsored ads overlap for purchasers in the last 30 days?" - "Build me a path-to-purchase view: which ad types did buyers touch, in what order?" ## Notes - Everything here is read-only. Running a query changes nothing in your ad account. - Results are aggregated by the AMC clean-room rules. Very small groups are suppressed by Amazon, so tiny segments can come back empty; that is AMC working as designed. ## Related - [Install the plugin](/knowledge-base/plugin/getting-started/install-the-plugin) - [Skill catalog](/knowledge-base/plugin/reference/skill-catalog) - [Plugin FAQ](/knowledge-base/plugin/faq/faq) --- ### Run unattended and scheduled tasks URL: https://mixshift.ai/knowledge-base/plugin/how-to/run-unattended-and-scheduled-tasks Markdown: https://mixshift.ai/api/knowledge-base/md/plugin/how-to/run-unattended-and-scheduled-tasks Type: how-to Product: plugin Last modified: 2026-07-31T00:00:00.000Z Set up a service credential with a one-time setup code, then let the scheduled-task skill build a job that keeps working with nobody at the keyboard. # Run unattended and scheduled tasks Personal sign-in is for a person at the keyboard. Anything that runs on its own (a morning health check, a nightly data pull, a weekly report) uses a **service credential** instead, so it keeps working when nobody is around to click a sign-in link. ## The setup-code flow (recommended) You never handle a secret. The flow is built around a one-time code: 1. Your MixShift admin mints a **setup code** from the admin page and sends it to you. The code is single use and expires in ten minutes, so it is safe to paste in chat. 2. In a chat, say **"set up a service credential"** and paste the code when asked. 3. The plugin exchanges the code itself; the credential lands directly on the machine that will run the work. If you are the admin, ask us and we will point you at the minting page for your account. ## Let the skill build the whole task Say **"set up a scheduled mixshift task"** (or run `/mixshift-ai:mx-scheduled-task`). It owns the full lifecycle so the task survives restarts and fresh sessions: - picks a durable folder the task can always find again, - sets up the service credential inside it, - generates task instructions that re-find the credential and re-pull brand context at the start of every run, - verifies the first run end to end before calling it done. ## Good to know - Service credentials are read-only by default. If an unattended job should be able to change things in your ad account, that permission is granted explicitly when the credential is minted, and every change is still audited. - Scheduled analytical skills need brand context, and the generated task pulls it from your organization store on each run. Set the brand up once first: [Set up a brand](/knowledge-base/plugin/how-to/set-up-a-brand). ## If a scheduled task breaks A task reporting **"not signed in"**, **"no credentials found"**, or **"missing brand context"** is the fix-it case for the same skill: say "repair my scheduled mixshift task" and it re-anchors the credential and context. ## Related - [Sign in to MixShift](/knowledge-base/plugin/getting-started/sign-in) (interactive sign-in) - [Set up a brand](/knowledge-base/plugin/how-to/set-up-a-brand) --- ### Set up a brand URL: https://mixshift.ai/knowledge-base/plugin/how-to/set-up-a-brand Markdown: https://mixshift.ai/api/knowledge-base/md/plugin/how-to/set-up-a-brand Type: how-to Product: plugin Last modified: 2026-08-21T00:00:00.000Z Brand setup builds the context the analytical skills read, one time per brand. What it collects, what it produces, and how to run it. # Set up a brand Brand setup teaches the plugin how one of your brands actually works: what it sells, how the account is structured, what counts as branded traffic, and what a normal week looks like. The analytical skills (health checks, search-term reviews, bid reviews, monthly reports) read that brand context to make judgment calls instead of generic ones. You run it once per brand, and it gets sharper over time as you correct it. ## When you need it Exploration and reporting work with sign-in alone. The analytical skills need brand context first; if you ask for one before setup, the plugin tells you and offers to start. The [skill catalog](/knowledge-base/plugin/reference/skill-catalog) marks exactly which skills need it. ## Run it Say **"set up brand context for [brand]"** in a chat, or run `/mixshift-ai:mx-brand-context [brand]`. The skill walks through a structured setup: it collects account details you confirm along the way, runs a set of database queries against your data, and turns the result into a brand profile. Setup starts by checking what MixShift already knows. If your team set this brand up on another machine, it offers to adopt that work instead of re-asking the same questions. Values already configured in the platform are shown for you to confirm rather than asked from scratch, and the account's existing portfolio names are read for structure hints. Only genuinely new questions are asked cold. ## What you get - A typed brand profile the skills read on every run. - A plain-language narrative describing the brand and how to manage it. - A reviewable Brand Context page so a human can check what the plugin believes and correct it. Corrections stick: when you adjust a threshold or fix a wrong assumption, later runs use the corrected version. ## Good to know - Expect a working session, not an instant command; the skill asks real questions about the brand. - Shared brands publish their context to your organization store, and it arrives on its own: the first time a teammate works that account anywhere, whether a new machine or a fresh Cowork session, the brand's context shows up in the background. Nothing to pull, nothing to remember. - Signing in shows how many brands your org has set up and how many are not yet on that machine, so a new teammate can see right away what they are missing. - If a sync cannot reach the org store, the plugin says so instead of staying quiet, and confirms your local copy is unchanged. - Setup is per brand. An agency running many accounts sets up each brand it manages. ## Related - [Skill catalog](/knowledge-base/plugin/reference/skill-catalog) (which skills need brand context) - [Run a skill](/knowledge-base/plugin/how-to/run-a-skill) --- ### Skill catalog URL: https://mixshift.ai/knowledge-base/plugin/reference/skill-catalog Markdown: https://mixshift.ai/api/knowledge-base/md/plugin/reference/skill-catalog Type: reference Product: plugin Last modified: 2026-08-28T00:00:00.000Z Every MixShift skill, grouped by what you are trying to get done, with what each needs before it can run. # Skill catalog The current skill library, grouped the same way the in-chat guide groups it (say "mixshift help" to see this map in a chat). Every skill runs by name in plain English or directly as `/mixshift-ai:mx-[skill]`. Skills marked **needs brand setup** read your brand context; run [brand setup](/knowledge-base/plugin/how-to/set-up-a-brand) once per brand first. Everything else works with sign-in alone. Each skill links to a detail page with what it produces and how to run it. ## Get set up and signed in - [mx-welcome](/plugin-and-mcps/mx-welcome): first-run orientation, shows where you stand and what to do next. - [mx-auth-login](/plugin-and-mcps/mx-auth-login): browser sign-in, account switching, session refresh. - [mx-auth-service-setup](/plugin-and-mcps/mx-auth-service-setup): service credential for unattended runs, via a one-time setup code. - [mx-scheduled-task](/plugin-and-mcps/mx-scheduled-task): builds a scheduled task end to end and verifies its first run. - [mx-brand-context](/plugin-and-mcps/mx-brand-context): brand setup; builds the context the analytical skills read. - [mx-update](/plugin-and-mcps/mx-update): guided update plus "what's new" catch-up. ## Explore data and pull reports - [mx-data-explore](/plugin-and-mcps/mx-data-explore): query, sample, and CSV-export your MixShift warehouse (Sponsored Ads, DSP, Seller and Vendor Central revenue, orders, inventory, catalog). Read-only. - [mx-amazon-report](/plugin-and-mcps/mx-amazon-report): on-demand reports from Amazon for any window (orders, listings, FBA, Sales and Traffic, Brand Analytics, settlement, vendor). Read-only. - [mx-amazon-retail](/plugin-and-mcps/mx-amazon-retail): live retail lookups (catalog, inventory, orders, finances, pricing, listings). Read-only. ## Daily and weekly account health - [mx-portfolio-quick-scan](/plugin-and-mcps/mx-portfolio-quick-scan): one green/yellow/red card per account; do I need to log in today? **Needs brand setup.** - [mx-daily-health-check](/plugin-and-mcps/mx-daily-health-check): daily exception-based review of spend and efficiency anomalies. **Needs brand setup.** - [mx-runaway-spend-check](/plugin-and-mcps/mx-runaway-spend-check): catches keywords spiking or spending with zero conversions. **Needs brand setup.** - [mx-keyword-bid-health](/plugin-and-mcps/mx-keyword-bid-health): weekly bid review; pullback candidates and scale opportunities. **Needs brand setup.** ## Search terms, negatives, and harvesting - [mx-search-term-negation](/plugin-and-mcps/mx-search-term-negation): systematic irrelevance review and surgical negative keywords. **Needs brand setup.** - [mx-search-term-harvest](/plugin-and-mcps/mx-search-term-harvest): promotes terms that convert efficiently under auto or broad match into explicit targeting. **Needs brand setup.** - [mx-phrase-negative-discovery](/plugin-and-mcps/mx-phrase-negative-discovery): finds phrase-level negative candidates from your search-term corpus; every candidate is validated before use. **Needs brand setup.** - [mx-asin-target-negation](/plugin-and-mcps/mx-asin-target-negation): reviews ASIN targets matched through auto and category paths for product-page overlap. **Needs brand setup.** - [mx-ppc-relevance-check](/plugin-and-mcps/mx-ppc-relevance-check): semantic relevance verdicts on search terms, separate from spend thresholds. **Needs brand setup.** - [mx-search-term-data-pull](/plugin-and-mcps/mx-search-term-data-pull): the data-extraction layer the other search-term skills build on. **Needs brand setup.** ## Live Amazon surfaces - [mx-amazon-ads](/plugin-and-mcps/mx-amazon-ads): live Ads account reads (campaigns, bids, budgets, recommendations) and audited writes. Changes preview first and never reach Amazon without your confirmation. - [mx-amazon-amc](/plugin-and-mcps/mx-amazon-amc): Amazon Marketing Cloud clean-room analytics in plain English. See [Run AMC queries from Claude](/knowledge-base/plugin/how-to/run-amc-queries-from-claude). Read-only. - [mx-amazon-dsp](/plugin-and-mcps/mx-amazon-dsp): DSP report pulls (campaign, inventory, audience), plus reads of how a DSP account is actually set up: campaigns, ad groups, creatives, which creative is attached to which line item, and whether a placement is running, paused, or was rejected and why. Straight from Amazon. Read-only. ## Reporting and analysis - [mx-monthly-report](/plugin-and-mcps/mx-monthly-report): the monthly performance report with month-over-month, year-over-year, efficiency analysis, and a looking-ahead section. **Needs brand setup.** ## Help, feedback, and contributing - [mx-help](/plugin-and-mcps/mx-help): the in-chat capability map and router when you are stuck. - [mx-feedback](/plugin-and-mcps/mx-feedback): sends a bug report, feature request, or comment straight to the MixShift team. - [mx-share-skill](/plugin-and-mcps/mx-share-skill): submit a skill you built for review and possible inclusion in the shared library. ## Related - [What the plugin can do](/knowledge-base/plugin/reference/what-the-plugin-can-do) - [Run a skill](/knowledge-base/plugin/how-to/run-a-skill) - [Set up a brand](/knowledge-base/plugin/how-to/set-up-a-brand) --- ### What the plugin can do URL: https://mixshift.ai/knowledge-base/plugin/reference/what-the-plugin-can-do Markdown: https://mixshift.ai/api/knowledge-base/md/plugin/reference/what-the-plugin-can-do Type: reference Product: plugin Last modified: 2026-08-28T00:00:00.000Z The plugin's capability areas and its honest boundaries: what is read-only, what can write, and what is gated behind brand setup. # What the plugin can do The plugin covers six capability areas. Everything is included with your MixShift subscription; the plugin is in beta. ## Explore your warehouse Ask questions against your MixShift data warehouse in plain English: Sponsored Products, Brands, and Display, DSP, Seller Central and Vendor Central revenue and orders, inventory, and catalog. Results export to CSV on request, and large results save as files automatically. The warehouse connection is read-only at the database level. ## Pull reports from Amazon, on demand Request reports straight from Amazon for any merchant and window you are authorized for: orders, listings, FBA inventory and fees, returns, settlement, Sales and Traffic, Brand Analytics (including Search Query Performance), and Vendor Central reports. Useful when you need data MixShift does not already hold, or a known report for an ad-hoc time frame. ## Look up live retail state Fast point-in-time answers about your catalog, inventory, orders, finances, pricing, and listings, served live from Amazon rather than from a generated report. ## Run clean-room and DSP analytics Amazon Marketing Cloud queries in plain English with CSV results ([how it works](/knowledge-base/plugin/how-to/run-amc-queries-from-claude)), DSP report pulls, and DSP account reads for the accounts you are connected to: which campaigns and creatives exist, which creative is attached to which line item, and whether a placement is running, paused, or was rejected and why. All read-only; changing a DSP creative or placement still happens in the Amazon DSP console. ## Manage advertising, with guardrails Live reads of campaigns, bids, budgets, and recommendations, plus the analytical tier: daily health checks, runaway-spend detection, bid reviews, search-term negation and harvesting, portfolio triage, and the monthly report. Analytical skills read your brand context; see [Set up a brand](/knowledge-base/plugin/how-to/set-up-a-brand). Writes exist and are deliberately narrow: advertising changes (pause and enable, bids, budgets, keywords, targets, negatives) preview first, show you the exact change set, and reach Amazon only after you confirm. Every change is recorded in an audit trail. If you decline a preview, nothing is sent. ## Remember your brand A brand timeline records events that matter (launches, price changes, stockouts) so reports can explain movements instead of just charting them. Brand context publishes to your organization store automatically, from a brand's first setup onward rather than only once someone has shared it by hand, and the events recorded on a brand travel to the timeline with it, so the whole team works from the same memory. Every automatic share prints a notice in the chat, and automatic publishing can be turned off if you would rather share on request. Arrival is automatic too: the first time anyone on the team works an account that was set up elsewhere, its context reaches that machine in the background, with no pull step to remember. ## The boundaries, plainly - **The warehouse is read-only. Always.** The database user cannot write, so no chat instruction can change warehouse data. - **Advertising is the only surface that writes to Amazon**, and every write is preview-gated and audited. All other Amazon surfaces in the plugin are read-only. - **Analytical skills need brand setup first.** Exploration and reports work with sign-in alone. - **Everyone in your MixShift org sees the same brands.** Per-user data scoping is not available today. - **One MixShift organization per sign-in.** An agency working across several MixShift customer accounts switches by signing in to the other account. - **MixShift Intelligence is in closed beta.** Contact support@mixshift.io if you want early access. - **During beta, usage telemetry is attributed** (account and person) so we can fix what breaks quickly. The plugin repository documents exactly what is and is not collected, and how to opt out. ## Related - [Skill catalog](/knowledge-base/plugin/reference/skill-catalog) - [Run a skill](/knowledge-base/plugin/how-to/run-a-skill) - [Builder Platform](/knowledge-base/builder-platform) (the same data from your own code and other AI tools) --- ### Plugin FAQ URL: https://mixshift.ai/knowledge-base/plugin/faq/faq Markdown: https://mixshift.ai/api/knowledge-base/md/plugin/faq/faq Type: faq Product: plugin Last modified: 2026-07-31T00:00:00.000Z Common questions about the MixShift AI plugin: cost, where it runs, team access, agencies, writes, and data safety. # Plugin FAQ ## Does the plugin cost extra? No. It is included with your MixShift subscription. ## Where does the plugin run? The full experience (skills plus the bundled CLI) runs in Claude Code and in the Cowork desktop app, on Windows, Mac, and Linux, and admins can roll it out organization-wide. Other tools (ChatGPT, Cursor, Codex, claude.ai connectors) reach the same data over MCP through the [Builder Platform](/knowledge-base/builder-platform); they get the data connection, not the packaged skills. ## Can the plugin change things in my ad account? Not without you seeing it first. Advertising is the only surface that can write to Amazon at all, and every change runs as a preview: you see the exact change set and confirm before anything reaches Amazon. Declined previews send nothing. Every applied change lands in an audit trail. The warehouse itself is read-only at the database level. ## Do I need my own Amazon developer accounts? No. Your Amazon accounts are already authorized through MixShift, so you skip the developer applications entirely and there is no token plumbing to build or maintain. ## Who at my company can see what? Everyone at your MixShift org sees the same brands once signed in. Per-user data scoping is not available today. ## We are an agency with several MixShift accounts. How do we switch? One sign-in maps to one MixShift organization. To work in another account, sign in again with that account's email. Smoother multi-account switching is on the roadmap. ## Is my data safe? You sign in on the MixShift sign-in page; the plugin never sees your password and only holds short-lived tokens that refresh automatically. The warehouse connection cannot write. During beta, usage telemetry is attributed (account and person) so we can fix problems fast; the plugin repository documents exactly what is collected and how to opt out. ## Why does a skill say it needs brand context? The analytical skills make brand-specific judgment calls, so they read a brand profile you build once per brand. It takes one working session and gets sharper as you correct it. See [Set up a brand](/knowledge-base/plugin/how-to/set-up-a-brand). ## How do I keep it current? Say "what's new in mixshift?" any time, and "update the mixshift plugin" when you want to take the update. One rule: a running session keeps its version, so restart after updating. See [Keep the plugin updated](/knowledge-base/plugin/how-to/keep-the-plugin-updated). ## How do I get help or report a problem? Say "mixshift help" for the capability map, or "send feedback to mixshift" to file a bug, request, or comment straight to the team from the chat. You can also write support@mixshift.io. ## Related - [Install the plugin](/knowledge-base/plugin/getting-started/install-the-plugin) - [What the plugin can do](/knowledge-base/plugin/reference/what-the-plugin-can-do) - [Troubleshooting](/knowledge-base/plugin/troubleshooting/troubleshooting) --- ### Troubleshooting URL: https://mixshift.ai/knowledge-base/plugin/troubleshooting/troubleshooting Markdown: https://mixshift.ai/api/knowledge-base/md/plugin/troubleshooting/troubleshooting Type: troubleshooting Product: plugin Last modified: 2026-07-31T00:00:00.000Z The plugin problems people actually hit, and the fix for each: sign-in, installs on Windows, updates that don't load, and results that arrive as files. # Troubleshooting The problems people actually hit, with the fix for each. If yours is not here, say **"send feedback to mixshift"** in the chat and describe it; that reaches the team directly. ## "Not signed in" or brands stop loading Sessions refresh automatically, but a long-idle machine can drop one. Say **"sign in to mixshift"** and run the 30-second flow again. If a scheduled task (not a person) hit this, see [Run unattended and scheduled tasks](/knowledge-base/plugin/how-to/run-unattended-and-scheduled-tasks); the fix-it path there re-anchors its credential. ## Install fails or loops on Windows Antivirus scanners can briefly lock files mid-install and the marketplace add fails partway. Retry the install; it succeeds on a clean pass. If it keeps happening, add an exclusion for your Claude plugin folder and retry. ## Commands will not run at all - Check Node.js: run `node --version` in a terminal; the plugin needs version 20 or newer. - On a personal Claude plan, make sure code execution is turned on under claude.ai Settings, Capabilities. ## I updated but nothing changed A running session keeps the version it started with, and starting a new chat or session does not pick up the new one. Fully quit the application and relaunch it. If the old version still shows after that, an earlier copy of the application is usually still running in the background and still serving the old build; close that one too. Asking to update the plugin will report the build this session is running alongside the build installed on your machine, which tells you whether the update landed or only the relaunch is outstanding. In Cowork, if the plugin panel still shows an old version label afterward, remove and re-add the marketplace entry; your sign-in and settings survive. ## My query result arrived as a file instead of in chat That is deliberate. Large results save as CSV files with a short preview so the conversation stays usable and nothing is truncated. Ask for the file contents inline if you really want them in chat. ## A report pull seems stuck Some Amazon report types generate slowly on the Amazon side (AMC especially: minutes, not seconds). The skill keeps polling and reports when done. Let it finish; cancelling and resubmitting just restarts the queue. ## A skill says it needs brand context That skill reads a brand profile. Run [brand setup](/knowledge-base/plugin/how-to/set-up-a-brand) once for that brand and re-run. ## Still stuck? - "mixshift help" routes you from inside the chat. - "send feedback to mixshift" files the problem with the team. - Or email support@mixshift.io. ## Related - [Plugin FAQ](/knowledge-base/plugin/faq/faq) - [Keep the plugin updated](/knowledge-base/plugin/how-to/keep-the-plugin-updated) --- ### Report Center 2.0 URL: https://mixshift.ai/knowledge-base/report-center Markdown: https://mixshift.ai/api/knowledge-base/md/report-center Type: index Product: report-center Last modified: 2026-06-15T00:00:00.000Z The MixShift analytics dashboard for Amazon retail and advertising performance. # Report Center 2.0 Report Center is the MixShift dashboard for tracking how your Amazon business is doing. It brings your retail sales and your advertising performance together in one place so you can see revenue, spend, ACoS, ROAS, and more across every marketplace you sell in. It works for both Seller and Vendor accounts, and it is built for brands, agencies, and merchants who manage Amazon advertising. ## Start here - [Getting started with Report Center](/knowledge-base/report-center/getting-started/getting-started-with-report-center): create your account, connect Amazon, and open your first report. - [What Report Center can do](/knowledge-base/report-center/reference/what-report-center-can-do): a tour of the features available today. ## How-to guides - [Build and read a report](/knowledge-base/report-center/how-to/build-and-read-a-report) - [Share a report](/knowledge-base/report-center/how-to/share-a-report) - [Export a report (CSV, Excel, PDF)](/knowledge-base/report-center/how-to/export-a-report) - [Schedule a report](/knowledge-base/report-center/how-to/schedule-a-report) - [Invite and manage your team](/knowledge-base/report-center/how-to/manage-your-team) - [Work across multiple organizations](/knowledge-base/report-center/how-to/work-across-multiple-organizations) ## More - [Frequently asked questions](/knowledge-base/report-center/faq/faq) - [Troubleshooting](/knowledge-base/report-center/troubleshooting/troubleshooting) - [Why don't my numbers match Seller Central?](/knowledge-base/report-center/troubleshooting/numbers-dont-match-seller-central) - [Contact support](/knowledge-base/shared/how-to/contact-support) --- ### Getting started with Report Center URL: https://mixshift.ai/knowledge-base/report-center/getting-started/getting-started-with-report-center Markdown: https://mixshift.ai/api/knowledge-base/md/report-center/getting-started/getting-started-with-report-center Type: getting-started Product: report-center Last modified: 2026-07-21T00:00:00.000Z Create your account, connect your Amazon data, and open your first report. # Getting started with Report Center This guide takes you from sign up to your first report. The whole setup takes a few minutes, and your data starts flowing right after you connect Amazon. ## 1. Create your account Sign up with your email and a password, or use "Continue with Google." If you sign up with email, check your inbox and click the verification link to confirm your address. ## 2. Set up your profile Add your name and time zone. Report Center uses your time zone so dates and schedules line up with your day. A photo is optional. ## 3. Create your organization Enter your company name and your role. This creates your organization, and you become its Owner. You can invite teammates later, and you can belong to more than one organization if you work across companies. ## 4. Connect Amazon (Selling Partner API) Click to connect, and you will be sent to Amazon to authorize MixShift. This is what lets Report Center pull your sales, orders, and financial data. Your connection is stored securely. ## 5. Connect Amazon Advertising (optional, recommended) You can also connect the Amazon Advertising API. This is optional, but we recommend it: without it, your reports will not include advertising metrics like spend, ACoS, and ROAS. You can skip this step now and connect later from Settings. ## 6. Open your first report When setup is done, go to the Reports page. Report Center automatically creates a set of default reports for each brand you connected, broken out by marketplace. Click into one to start exploring. ## A note on your first day Your Amazon data syncs in the background. If you open Report Center and a report looks empty, you may see a banner letting you know your data is still syncing. Data refreshes once per day, so if you just connected, give it a little time and check back. You will get an email letting you know when your dashboard is ready, so you do not need to keep checking. ## What's next - [Build and read a report](/knowledge-base/report-center/how-to/build-and-read-a-report) - [What Report Center can do](/knowledge-base/report-center/reference/what-report-center-can-do) --- ### Build and read a report URL: https://mixshift.ai/knowledge-base/report-center/how-to/build-and-read-a-report Markdown: https://mixshift.ai/api/knowledge-base/md/report-center/how-to/build-and-read-a-report Type: how-to Product: report-center Last modified: 2026-06-15T00:00:00.000Z Open a report, set your date range and filters, and read the results. # Build and read a report Report Center creates reports for you automatically, so building one is really about shaping the view to answer your question. Here is how. ## 1. Open a report On the Reports page, find the brand you want and click into one of its reports. Use search at the top if you have a lot of brands, and star the ones you open often so they stay at the top. ## 2. Set your date range Use the date picker to choose a preset like Last 30 Days or Year to Date, or set a custom range. Then choose how to group the data with the day, week, or month toggle. ## 3. Compare against a baseline Turn on a comparison to see whether you are up or down. You can compare to the prior period (for example, the previous 30 days) or the prior year. Your KPI scorecards will update to show the change. ## 4. Filter to what matters Use filters to narrow the report by brand, item group, or marketplace. Active filters show as chips at the top, and you can clear them one at a time or all at once. ## 5. Read the results - **KPI scorecards** at the top show your headline metrics, with the change versus your comparison shown as a number and a percentage. - **The trend chart** shows those metrics over time. Hover any point for detail. - **The item breakdown table** shows the same metrics by ASIN, SKU, item nickname, brand, or item group. Use the "View by" selector to switch dimensions, sort by any column, and drag columns to reorder them. - **Hover an ASIN** to see its name, image, and details, or click it to open the product on Amazon. ## 6. Choose your metrics Use the metric picker to add or remove metrics. Report Center only shows metrics that are valid for the level you are viewing, so the list changes as you switch dimensions. ## Notes - Data refreshes once per day, so the latest numbers reflect your most recent nightly sync. - Your date range and filters are saved in the page link, so you can copy the URL to return to the exact same view. ## Related - [Share a report](/knowledge-base/report-center/how-to/share-a-report) - [Export a report](/knowledge-base/report-center/how-to/export-a-report) --- ### Export a report (CSV, Excel, PDF) URL: https://mixshift.ai/knowledge-base/report-center/how-to/export-a-report Markdown: https://mixshift.ai/api/knowledge-base/md/report-center/how-to/export-a-report Type: how-to Product: report-center Last modified: 2026-07-21T00:00:00.000Z Download a report as CSV, Excel, or PDF, matching your current view. # Export a report (CSV, Excel, PDF) You can download a report to share offline or work with the numbers elsewhere. Exports match what you see on screen, including your active filters, date range, and view. ## Export to CSV 1. Open the report and set your date range, filters, and "View by" the way you want them. 2. Choose Export and select CSV. 3. The file downloads with the merchant, report type, and date range in the filename. CSV export is the most reliable option and is best for production workflows. ## Export to Excel (Beta) Choose Export and select Excel to download a workbook with separate tabs for your KPIs, item breakdown, and time series, with numbers and currency formatted for you. Excel export is in beta. It works well for typical reports, but very large or complex reports may hit edge cases. If you need a guaranteed result, use CSV. ## Export to PDF (Beta) Choose Export and select PDF to download a print-ready version of the report as you see it on screen, including the report details and a generated timestamp. PDF export is in beta. Today the download arrives as a web page file (.html) rather than a finished PDF: open it in your browser, choose Print, and save as PDF. Very long reports or complex charts may not render perfectly. ## Notes - Exports respect your current filters and date range, so set your view before exporting. - If you need exports delivered on a schedule, see [Schedule a report](/knowledge-base/report-center/how-to/schedule-a-report). ## Related - [Build and read a report](/knowledge-base/report-center/how-to/build-and-read-a-report) - [Schedule a report](/knowledge-base/report-center/how-to/schedule-a-report) --- ### Invite and manage your team URL: https://mixshift.ai/knowledge-base/report-center/how-to/manage-your-team Markdown: https://mixshift.ai/api/knowledge-base/md/report-center/how-to/manage-your-team Type: how-to Product: report-center Last modified: 2026-07-21T00:00:00.000Z Invite teammates by email, assign roles, and manage access in your organization. # Invite and manage your team Bring teammates into your organization and control what each person can do. ## Invite a teammate 1. Go to the Team page. 2. Enter the person's email address. 3. Choose a role (see below). 4. Send the invite. They will get an email with a link to accept. If they do not have a MixShift account yet, they will be prompted to create one. ## Roles - **Owner**: full control of the organization, including team and connections. - **Admin**: manage reports and team members. - **Member**: work with reports. You can change anyone's role from the Team page. When you invite someone, you can grant them the Admin or Member role. ## Remove a teammate Remove a person from the Team page. You will be asked to confirm. Note that any share links they created stay active until they are revoked, so revoke those links as well if you need to fully cut off access. ## Notes - Roles apply within an organization. If you work across multiple organizations, each has its own team and roles. - A history log of team changes is not available yet. ## Related - [Share a report](/knowledge-base/report-center/how-to/share-a-report) - [Getting started](/knowledge-base/report-center/getting-started/getting-started-with-report-center) --- ### Schedule a report URL: https://mixshift.ai/knowledge-base/report-center/how-to/schedule-a-report Markdown: https://mixshift.ai/api/knowledge-base/md/report-center/how-to/schedule-a-report Type: how-to Product: report-center Last modified: 2026-06-15T00:00:00.000Z Set a report to send automatically on a daily, weekly, or monthly cadence. # Schedule a report > **Beta.** Scheduling is available, but automated delivery is still being finalized. Please read the limitations below before relying on it. You can set a report to generate on a recurring schedule and send to a list of recipients. ## Create a schedule 1. Open the report you want to send on a schedule. 2. Go to Scheduled Reports and create a new schedule. 3. Choose the cadence (daily, weekly, or monthly) and the time. 4. Add your recipients. 5. Choose the format (CSV, Excel, or PDF). 6. Save the schedule. ## Manage your schedules On the Scheduled Reports page you can see all your schedules, their last run, and when they run next. You can edit, pause, or delete any schedule. ## Current limitations (Beta) While scheduling is in beta: - **Automated email delivery is still being wired up**, so scheduled sends may not arrive reliably yet. - **The PDF format for scheduled reports is not final.** For dependable output today, use CSV or export manually. If you need a report right now, the most reliable path is to open it and [export it manually](/knowledge-base/report-center/how-to/export-a-report). We will update this guide as scheduled delivery moves out of beta. ## Related - [Export a report](/knowledge-base/report-center/how-to/export-a-report) - [Build and read a report](/knowledge-base/report-center/how-to/build-and-read-a-report) --- ### Share a report URL: https://mixshift.ai/knowledge-base/report-center/how-to/share-a-report Markdown: https://mixshift.ai/api/knowledge-base/md/report-center/how-to/share-a-report Type: how-to Product: report-center Last modified: 2026-06-15T00:00:00.000Z Create a secure share link, protect it with a password, and control who can view or edit. # Share a report You can share any report with teammates or clients, with or without a MixShift account, and you stay in control of access. ## Create a share link 1. Open the report you want to share. 2. Open the share options and create a share link. 3. The link is copied to your clipboard, ready to send. Anyone with the link can open the report, even without a MixShift account. ## Protect a link with a password 1. When you create the link, turn on password protection and set a password. 2. Share the password with your recipients separately. 3. They will be asked for the password before they can view the report. ## Choose view-only or editable - **View only** lets recipients see the report exactly as you shared it. - **Editable** lets them change the date range and filters to explore on their own. Your original is not affected. ## Share several reports at once 1. On the Reports page, select multiple reports. 2. Apply a single access level and recipient list. 3. All selected reports are shared in one step. ## Turn off access You can revoke a link at any time. Access ends immediately, and the next time someone opens the link they will see that the report is no longer available. ## Notes - Reports that others have shared with you are marked clearly in your report list. - A log of who opened a shared link is not available yet, and links do not expire automatically. To end access, revoke the link. ## Related - [Invite and manage your team](/knowledge-base/report-center/how-to/manage-your-team) - [Export a report](/knowledge-base/report-center/how-to/export-a-report) --- ### Work across multiple organizations URL: https://mixshift.ai/knowledge-base/report-center/how-to/work-across-multiple-organizations Markdown: https://mixshift.ai/api/knowledge-base/md/report-center/how-to/work-across-multiple-organizations Type: how-to Product: report-center Last modified: 2026-07-21T00:00:00.000Z If you belong to more than one organization, Report Center shows one at a time. Switch from the organization menu. # Work across multiple organizations If you work across companies (for example, you are at an agency or you consult for several sellers), you can belong to more than one MixShift organization. Report Center shows one organization at a time: the one you are currently working in. Your reports, brands, and sellers are scoped to that active organization. ## Switch organizations 1. Open the organization menu in the header. 2. Pick the organization you want to work in. 3. Report Center reloads with that organization's brands and reports. If you belong to only one organization, there is no menu to switch. You only see it when you are a member of two or more. ## What switching changes - The Reports page shows only the active organization's brands and reports. - Sharing, exports, and schedules you create belong to the organization you created them in. - Your role can be different in each organization. You might be an Owner in one and a Member in another. ## Notes - You are invited into an organization by its Owner or an Admin. See [Invite and manage your team](/knowledge-base/report-center/how-to/manage-your-team). - If you expect to see a brand and it is missing, check which organization you are in first. That solves it most of the time. ## Related - [Invite and manage your team](/knowledge-base/report-center/how-to/manage-your-team) - [Getting started](/knowledge-base/report-center/getting-started/getting-started-with-report-center) --- ### What Report Center can do URL: https://mixshift.ai/knowledge-base/report-center/reference/what-report-center-can-do Markdown: https://mixshift.ai/api/knowledge-base/md/report-center/reference/what-report-center-can-do Type: reference Product: report-center Last modified: 2026-07-21T00:00:00.000Z A tour of the features available in Report Center today. # What Report Center can do A tour of what is available today. Everything below is live and ready to use, unless it is marked **Beta**. ## Reports and analysis - **Default reports for every brand.** When you connect Amazon, Report Center creates reports for each brand and marketplace automatically. The main types are Retail and Sponsored Ads Performance, Retail Performance, and Sponsored Ads Performance. - **Find reports fast.** The Reports page lists your brands with inline, searchable sections. Search by brand name, account ID, or report name, and star the brands you check most. - **Date ranges and comparisons.** Pick a preset like Last 30, 60, or 90 Days or Year to Date, or set a custom range. View by day, week, or month, and compare against the prior period or the prior year. - **KPI scorecards.** See your top metrics at a glance, with the change versus your comparison period shown as both a number and a percentage, color coded so wins and losses are obvious. - **Trend charts.** Track performance over time with interactive charts you can hover, zoom, and pan. - **Item breakdown table.** Break results down by ASIN, SKU, item nickname, brand, or item group. Sort and filter the rows, and reorder the metric columns to your liking. - **Monthly trend matrix.** See a rolling 12 months of any metric with month over month changes. - **Choose your metrics.** Pick which metrics show on a report. Report Center only offers metrics that make sense at the level you are viewing. - **Item details.** Hover an ASIN to see its name, parent ASIN, brand, item group, and a product image, and click through to the Amazon product page. ## Sharing and collaboration - **Share links.** Create a secure link to any report and copy it with one click. - **Password protection.** Add a password to a share link so only the right people can open it. - **View or edit access.** Decide whether someone can just view a report or also change filters and date ranges. - **Share in bulk.** Select multiple reports and share them all to the same people at once. - **Revoke anytime.** Turn off access instantly. The next time someone opens the link, they will see that it is no longer available. ## Exports - **CSV export.** Download any report as a CSV that respects your current filters, date range, and view. *(Live.)* - **Excel export. (Beta)** Download a multi-tab Excel file with your KPIs, item breakdown, and time series. - **PDF export. (Beta)** Download a print-ready copy of your report. Today it arrives as a web page file you print to PDF from your browser. ## Scheduling - **Scheduled reports. (Beta)** Set a report to send on a daily, weekly, or monthly cadence to a list of recipients. See [Schedule a report](/knowledge-base/report-center/how-to/schedule-a-report) for current limitations. ## Your team and account - **Invite teammates** by email and give each person a role: Owner, Admin, or Member. - **Manage connections.** Review your connected Amazon accounts, see the last sync time, and reconnect if a connection expires. - **Profile and password.** Update your name, time zone, and password from Settings. - **Data Manager (browse).** Browse your campaign and item metadata in a filterable table. ## Good to know - **Data refreshes once per day.** Report Center syncs your Amazon data nightly. Real-time data is not available yet. - Some areas you may have used in MixShift 1.0 are not in Report Center yet. See [the FAQ](/knowledge-base/report-center/faq/faq) for what is and is not available today. ## Related - [Getting started](/knowledge-base/report-center/getting-started/getting-started-with-report-center) - [Build and read a report](/knowledge-base/report-center/how-to/build-and-read-a-report) --- ### Report Center FAQ URL: https://mixshift.ai/knowledge-base/report-center/faq/faq Markdown: https://mixshift.ai/api/knowledge-base/md/report-center/faq/faq Type: faq Product: report-center Last modified: 2026-07-21T00:00:00.000Z Common questions about data, accounts, sharing, and what is available today. # Report Center FAQ ## How often does my data update? Once per day. Report Center syncs your Amazon data nightly, so your reports reflect your most recent sync. Real-time data is not available yet. ## Do I have to connect Amazon Advertising? No, but we recommend it. Without the Advertising connection, your reports will not include advertising metrics like spend, ACoS, and ROAS. You can connect it any time from Settings. ## Can I share a report with someone who does not use MixShift? Yes. Create a share link and send it. They can open the report without an account. You can add a password and choose whether they can view only or also adjust filters and dates. ## Can I export my data? Yes. CSV export is fully available. Excel export is in beta and works well for most reports. PDF export is in beta and currently downloads as a web page you print to PDF from your browser. See [Export a report](/knowledge-base/report-center/how-to/export-a-report). ## Can I schedule reports to send automatically? Scheduling is in beta. You can create and manage schedules, but automated email delivery is still being finalized, so it is not fully reliable yet. See [Schedule a report](/knowledge-base/report-center/how-to/schedule-a-report). ## Does Report Center support Vendor accounts? Yes. Report Center supports both Seller and Vendor accounts. The combined Retail and Sponsored Ads Performance report is available for Vendor accounts today. A dedicated Vendor Retail Performance view with traffic metrics like Sessions and Page Views is planned but not available yet. ## I used a feature in MixShift 1.0 that I cannot find. Where is it? Report Center is still growing toward full parity with 1.0. Some areas are not available yet, including: Keywords and Targets, Budget Balancing, Pacing, the Data Master drill-down explorer, settlement reports, and the legacy Merchant Dashboard. If you depend on one of these, let us know so we can prioritize. ## Is there an AI assistant or automatic insights? Not yet. Automatic narratives and anomaly explanations are in development and are not available today. Your reports give you the metrics, charts, and breakdowns to analyze performance yourself. ## Does Report Center have two-factor authentication or billing built in? Not yet. Two-factor authentication and in-app billing are planned but not available today. ## My numbers look different from Seller Central. Is something broken? Usually not; small differences on recent days are expected. Amazon revises recent days after the fact, ad conversions get credited to the click day, and MixShift syncs once per day. See [Why don't my numbers match Seller Central?](/knowledge-base/report-center/troubleshooting/numbers-dont-match-seller-central) for how to compare fairly. ## How do I get help? Email support@mixshift.io or use the Contact page on mixshift.ai. See [Contact MixShift support](/knowledge-base/shared/how-to/contact-support) for what to include so we can fix it on the first pass. ## Related - [What Report Center can do](/knowledge-base/report-center/reference/what-report-center-can-do) - [Troubleshooting](/knowledge-base/report-center/troubleshooting/troubleshooting) --- ### Report Center troubleshooting URL: https://mixshift.ai/knowledge-base/report-center/troubleshooting/troubleshooting Markdown: https://mixshift.ai/api/knowledge-base/md/report-center/troubleshooting/troubleshooting Type: troubleshooting Product: report-center Last modified: 2026-06-15T00:00:00.000Z Fixes for common issues with data, connections, and access. # Report Center troubleshooting ## My report is empty or missing data - If you just connected Amazon, your data may still be syncing. Look for a sync banner on the Reports page and check back a little later. Data syncs once per day. - Make sure the right Amazon connection is in place. Open Settings and confirm your Selling Partner API connection is active and has a recent sync time. - If you expected advertising metrics (spend, ACoS, ROAS) and do not see them, you may not have connected Amazon Advertising. Connect it from Settings. ## My numbers look out of date Report Center refreshes once per day. The latest figures reflect your most recent nightly sync. Real-time data is not available yet. ## A connection expired or stopped syncing Open Settings and check your connected Amazon accounts. If a connection shows an expired or revoked status, reauthorize it. Your data resyncs after you reconnect. ## I was signed out unexpectedly For your security, sessions expire after a period of inactivity. Sign back in, and you will be returned to the page you were on. ## A teammate still has access after I removed them Removing a teammate does not automatically turn off share links they created. Go to the report's share settings and revoke any links that should no longer work. ## My scheduled report did not arrive Scheduling is in beta and automated email delivery is still being finalized, so scheduled sends are not fully reliable yet. For now, open the report and [export it manually](/knowledge-base/report-center/how-to/export-a-report). ## Still stuck? Reach out to the MixShift team with your organization name and the report you were working on, and we will help. ## Related - [FAQ](/knowledge-base/report-center/faq/faq) - [Getting started](/knowledge-base/report-center/getting-started/getting-started-with-report-center) --- ### Why don't my numbers match Seller Central? URL: https://mixshift.ai/knowledge-base/report-center/troubleshooting/numbers-dont-match-seller-central Markdown: https://mixshift.ai/api/knowledge-base/md/report-center/troubleshooting/numbers-dont-match-seller-central Type: troubleshooting Product: report-center Last modified: 2026-07-21T00:00:00.000Z The four honest reasons your MixShift numbers can differ from what Amazon shows, and how to compare them fairly. # Why don't my numbers match Seller Central? You pull up a report, open Seller Central next to it, and the numbers disagree. Before assuming something is broken: small differences on recent days are normal, and almost always come down to one of four things. Here is how to tell which one you are looking at. ## 1. Your data is one day behind, on purpose MixShift syncs your Amazon data once per day. Seller Central shows you this moment. If you compare today (or late yesterday), Seller Central will be ahead. Compare completed days, not the current one. ## 2. Amazon revises recent days after the fact The numbers Amazon reports move for a while after the day ends. Orders get cancelled or refunded, and financial details settle over days, sometimes up to two weeks. MixShift re-pulls recent days on each sync to pick up those revisions, so a number you saw Monday can legitimately change by Thursday. Both were "right" when shown. If a recent day looks off, check it again after a few more days before treating it as a discrepancy. ## 3. Advertising attribution moves conversions backward An ad conversion is credited to the day of the click, not the day of the purchase. If someone clicks Tuesday and buys Friday, Amazon adds that sale to Tuesday, days later. Advertising metrics for the last several days therefore keep climbing after the fact. This is the attribution model Amazon uses, and it affects any tool reading the same data. ## 4. You are not comparing the same slice The most common cause, and the easiest to fix. Before comparing, make sure both sides use: - the same date range, in the same time zone, - the same marketplace (a single marketplace in Seller Central versus all marketplaces in a MixShift report is the classic mismatch), - the same metric definition. "Sales" in one surface can mean ordered product sales, and shipped revenue in another. ## How to compare fairly 1. Pick one completed day at least two weeks old. 2. Pick one marketplace, same on both sides. 3. Compare one specific metric (for example ordered product sales). If the numbers still disagree materially after that, something genuinely needs a look: check [Troubleshooting](/knowledge-base/report-center/troubleshooting/troubleshooting) for sync issues, or [contact support](/knowledge-base/shared/how-to/contact-support) with the day, marketplace, and metric you compared, and we will trace it. ## Related - [Report Center troubleshooting](/knowledge-base/report-center/troubleshooting/troubleshooting) - [How MixShift keeps your data fresh](/knowledge-base/data-hub/concept/how-mixshift-keeps-your-data-fresh) --- ### Account & Support URL: https://mixshift.ai/knowledge-base/shared Markdown: https://mixshift.ai/api/knowledge-base/md/shared Type: index Product: shared Last modified: 2018-10-20T01:46:40.000Z Cross-product help. Getting support, and answers that apply to every MixShift product. # Account & Support Help that applies across every MixShift product. ## How to - [Contact MixShift support](/knowledge-base/shared/how-to/contact-support) --- ### Contact MixShift support URL: https://mixshift.ai/knowledge-base/shared/how-to/contact-support Markdown: https://mixshift.ai/api/knowledge-base/md/shared/how-to/contact-support Type: how-to Product: shared Last modified: 2026-07-21T00:00:00.000Z How to reach a human, and what to include so we can fix your issue on the first pass. # Contact MixShift support Stuck, or found something that looks wrong? Reach us here: - **Email:** support@mixshift.io - **Contact form:** the Contact page on mixshift.ai (pick the product your question is about) ## What to include The fastest fixes come from specific reports. Include what you can of: 1. **Where you were.** The product and page (for example "Report Center, the export screen"). 2. **What you did, what you expected, what happened instead.** Three short sentences beat a paragraph. 3. **The specifics.** Brand, marketplace, date range, and metric if your question is about numbers. A screenshot helps more than a description. 4. **When it happened**, with your time zone. ## Data questions If your question is "this number looks wrong," first see [Why don't my numbers match Seller Central?](/knowledge-base/report-center/troubleshooting/numbers-dont-match-seller-central) It resolves most cases. If yours survives that article, send us the single day, marketplace, and metric you compared, and we will trace it through the pipeline. ## Related - [Report Center troubleshooting](/knowledge-base/report-center/troubleshooting/troubleshooting) --- ## Blog Content ### I called the Netflix Tyson fight failure. Amazon has the same problem every Prime Day URL: https://mixshift.ai/blog/live-event-scale-prime-day Date: 2026-06-11 Author: Tony Tags: data-infrastructure, amazon-practical I told my team Netflix was going to crash before the Tyson fight even started. Not because I'm clairvoyant. I'd lived the same problem at Pluto TV for years. And Amazon has it every Prime Day. --- ### How Todd Solved Forecasting for Amazon Agencies and Brands URL: https://mixshift.ai/blog/forecasting-solved-for-amazon-agencies-and-brands Date: 2025-10-15 Author: Mixshift Tags: amazon-forecasting A deep dive into the methodology and real-world use cases behind MixShift's revenue forecasting model for Amazon brands and agencies. --- ### Why BI Tools Break Down for an Amazon Agency's Client Reporting URL: https://mixshift.ai/blog/bi-tools-break-down-for-amazon-agency-client-reporting Date: 2025-06-06 Author: Mixshift Tags: amazon-reporting, amazon-agencies A detailed breakdown of why general-purpose BI tools fail to meet the specific needs of Amazon agencies managing client reporting at scale. --- ### Why Amazon Reporting Breaks at Scale (And What to Do About It) URL: https://mixshift.ai/blog/why-amazon-reporting-breaks-at-scale Date: 2025-05-29 Author: Mixshift Tags: amazon-reporting The 13 specific ways that manual spreadsheet-based reporting systems fail Amazon agencies as they grow their client rosters. --- ### Boosting Amazon Purchase Orders: How Agencies Can Help Vendors Get More Orders from Amazon URL: https://mixshift.ai/blog/help-vendors-get-more-purchase-orders-from-amazon Date: 2023-11-17 Author: Mixshift Tags: amazon-vendor-central An insider look at Amazon's automated inventory ordering system and the strategies agencies can use to help vendor clients secure more purchase orders. --- ### Why the SP-API Health Dashboard May Not Provide the Full Picture for Amazon Sellers URL: https://mixshift.ai/blog/why-the-sp-api-health-dashboard-may-not-provide-the-full-picture-for-amazon-sellers Date: 2023-08-10 Author: Mixshift Tags: sp-api Three critical limitations of Amazon's SP-API Health Dashboard that sellers and developers should understand before relying on it for outage detection. --- ### All About Amazon Advertising Billing – How Does It Work? URL: https://mixshift.ai/blog/all-about-amazon-advertising-billing Date: 2023-07-23 Author: Mixshift Tags: amazon-advertising A complete guide to how Amazon Advertising billing works, including payment methods, credit limits, invoice access, and where to manage your settings. --- ### How to Get Started with Amazon Ads API Reporting Data URL: https://mixshift.ai/blog/how-to-amazon-advertising-api-reports Date: 2023-06-30 Author: Mixshift Tags: amazon-advertising, amazon-api A practical guide to leveraging Amazon's advertising API reports for agencies, aggregators, sellers, and vendors building scalable reporting systems. --- ### How to Know When Amazon's Advertising Data Isn't Right URL: https://mixshift.ai/blog/how-to-know-when-amazons-advertising-data-isnt-right Date: 2023-05-19 Author: Mixshift Tags: amazon-advertising, amazon-reporting Why no data provider can independently verify Amazon Advertising API accuracy - and what to look for when reported data may be compromised. --- ### Important Updates for Amazon FBA Inventory In SP-API URL: https://mixshift.ai/blog/amazon-fba-inventory-updates-in-sp-api Date: 2023-01-15 Author: Mixshift Tags: sp-api, amazon-fba Amazon is deprecating six FBA inventory report endpoints at the end of January - here's what's changing, what the new ledger reports are missing, and how to adapt. --- ### Looking for Your Vendor or Seller Amazon Data? Here's a Breakdown of the Current Amazon APIs for 2023 URL: https://mixshift.ai/blog/looking-for-your-vendor-or-seller-amazon-data-heres-a-breakdown-of-the-current-amazon-apis-for-2023 Date: 2023-01-03 Author: Mixshift Tags: amazon-api, sp-api A comprehensive overview of Amazon's four primary APIs - SP-API, Advertising API, Vendor Central API/EDI, and Product Advertising API - for vendors and sellers seeking data access. --- ### How Does Amazon Search Work? The Amazon Game Explained URL: https://mixshift.ai/blog/how-does-amazon-search-work Date: 2022-01-24 Author: Mixshift Tags: amazon-advertising An explanation of how Amazon's search algorithm evolved, why PPC advertising is not optional, and how ad conversions influence organic rankings. --- ## Contact Information - Website: https://mixshift.ai - Email: hello@mixshift.io - Phone: 1-833-DASH-APP (1-833-327-4277) - LinkedIn: https://www.linkedin.com/company/mixshift - Twitter: https://twitter.com/mixshiftsoftware - Facebook: https://www.facebook.com/MixShift - GitHub: https://github.com/mixshift