# Automate Recharge work with your agents

Maplith maps how Recharge’s subscriptions, customers, charges and orders join your web shop, payments and CRM, so your agent knows where each thing is and how far to trust each join.

Share Recharge’s exports today; Maplith maps them from there.

| Way | Status | What it means |
|---|---|---|
| Export files | Available | CSV exports: processed charges and orders for the last three months; all subscriptions and customers. The path to start with. |

## Example flows

What starts the work, what your agent does with Maplith, and how it ends. Each flow uses the Recharge data and joins described below.

### A subscriber charged with no order to ship

**Trigger:** A subscriber was charged in Recharge, and no matching order shows up in Shopify to ship.

1. **Findings**: Where the last scan of the exports saw processed charges with no Shopify order, a finding names them, with its evidence and a recommendation.
2. **Route**: `route` from the Recharge charge to the Shopify order by the Shopify order id the charge carries, with how much Maplith trusts that hop.
3. **The map**: `get_object` on the charges says a charge carries a status, and that the order Recharge sends to the shop comes from a processed one. The agent looks for the order in Shopify with its own read access.
4. **The action gate**: `request_action` for a task for whoever runs fulfilment, with what the finding names and what the agent found in Shopify.
5. **Record and value**: `record_action` records the task, and whether the order was sent after all.

**Result:** A charge with nothing to ship reaches the person who runs fulfilment as one task, with the ids to look up.

### Subscriptions that renew on a retired variant

**Trigger:** A product variant is taken out of the Shopify catalogue, and subscriptions in Recharge may still renew on it.

1. **Ask**: `ask`: “how do Recharge subscriptions tie to Shopify’s products?” The answer cites the join on the Shopify variant id, and says where a SKU override can stand in for it.
2. **Route**: `route` from the Shopify variant to the Recharge subscriptions that carry its variant id.
3. **The map**: `get_object` on the subscriptions says which fields hold the Shopify variant id and the next charge date, so the ones on a variant can be found in Recharge by that id.
4. **The action gate**: `request_action` for a task for the person who manages subscriptions: the retired variant, its Shopify variant id, and the fields to find its subscriptions by in Recharge.
5. **Record and value**: `record_action` records that the task went out, and how it ended.

**Result:** The person who manages subscriptions knows which variant id to look for in Recharge before the next renewal runs.

### Recharge charges and payments that do not pair up

**Trigger:** At month-end, the subscription charges in Recharge and the payments at your payment provider do not agree.

1. **Findings**: Where the last scan of the exports found charges and payments that do not pair up, a finding with its evidence and a recommendation.
2. **Route**: `route` from a Recharge charge to the charge at Stripe, Braintree or another payment provider, by the transaction id the charge carries.
3. **The map**: `get_object` on the charges says a charge carries a status, so a queued or skipped charge is not compared with a payment. The agent reads the payments the finding names in Stripe with its own read access.
4. **The action gate**: `request_action` for a task for the bookkeeper with the finding, the payments the agent found, and the transaction ids to check in Recharge.
5. **Record and value**: `record_action` records that the task went to the bookkeeper, and what was left open.

**Result:** The bookkeeper sees which charges and payments do not pair up, by transaction id.

Steps that use `request_action` or `record_action` need an agent key that may act; a read-only key reads the map.

## How it connects

Read-only, through access you already have. Maplith holds no Recharge credential.

1. Export from Recharge: CSV, processed charges and orders for the last three months; all subscriptions and customers. The guide has the steps.
2. Upload the files in your workspace, or let your assistant send them to Maplith over MCP.
3. Maplith builds the map and deletes the files again.

### What Maplith never reads or does

- **Recharge itself:** Maplith holds no Recharge credential and never signs in to it.
- **Anything that changes Recharge:** Maplith never writes there.
- **What you leave out:** you decide what an export contains.

### Datasets that make a good map

- **Subscriptions**: One subscription: the product variant, quantity, schedule, status and next charge date. Key columns: subscription id, Recharge customer id, address id, external product id (Shopify product id), external variant id (Shopify variant id), SKU override, status, next charge date.
- **Customers**: One subscriber with active subscription count, total orders and payment method status. Key columns: Recharge customer id, external customer id (Shopify customer id), customer email (Maplith tries to mask it), number of active subscriptions.
- **Charges**: One scheduled or processed charge with its line items and status (queued, skipped, error, processed). Key columns: charge id, Recharge customer id, external order id (Shopify order id), payment processor transaction id, line item SKU, status, scheduled date.
- **Orders**: One processed order that a charge created and sent to the shop. Key columns: Recharge order id, charge id, external order id (Shopify order id), Recharge customer id, processed date.

### The narrowest access Recharge offers for reading

- `read_subscriptions`
- `read_customers`
- `read_orders`
- `read_products`
- `read_plans`
- `read_store`

The store owner; by default only store owners see the API tokens page, but they can give staff access to it.

Checked against Recharge’s documentation on 8 October 2026. Sources and export steps: [Read Recharge into Maplith](https://maplith.com/docs/md/systems/recharge).

## FAQ

### Does Maplith copy my Recharge data?

No. Maplith keeps metadata about your data: which datasets Recharge holds, their fields, and how they join your other systems. It does not keep copies of the records. What you share is read for a build and then deleted. The map may quote example values and evidence.

### Which Recharge access does Maplith need?

None: Maplith holds no Recharge credential. You share Recharge’s exports; the store owner can make them. If a tool of yours reads Recharge through an API token instead, the store owner creates one with Read access on each permission it needs, never Read and Write access.

### Which AI assistants work with Maplith and Recharge?

Any assistant that speaks MCP, such as Claude, ChatGPT or one you build, connects to Maplith by signing in or with an agent key. You share Recharge’s exports with Maplith, and your agent reads the map. [Connect your own agent](https://maplith.com/docs/own-agent)

### Where is my Recharge data processed?

In the EU, except model inference. Hosting, database and file storage are in the EU. Model inference runs in the United States under standard contractual clauses: to build the map, samples of the files and, in the findings step, whole files are sent to Maplith’s model provider, which deletes them within 30 days. [Security](https://maplith.com/security) states the details.

### Does Maplith change anything in Recharge?

No. Maplith holds no Recharge credential and never writes to Recharge: it reads only the exports you share. A change in Recharge stays with the people who work in Recharge.

### What about customer emails and other personal data in Recharge?

Recharge’s customers hold customer emails. Maplith tries to mask email local parts and the names in an export’s own name columns (on by default; an admin can switch it off). It is a best-effort safeguard, not a guarantee. You decide what an export contains.

### Does Recharge have an MCP server Maplith can read?

Not for a map. The Recharge Analytics MCP is for early adopters, needs a person’s own sign-in, and answers only totals such as subscriber counts and churn, never individual subscriptions or charges. The Export Builder gives the files with the Shopify ids that join them to your shop.

### Which Recharge data makes a good map?

Subscriptions, customers, charges and orders. For a first map, the exports cover processed charges and orders for the last three months; all subscriptions and customers.

## Related integrations

- [Shopify](https://maplith.com/integrations/md/shopify): From Shopify · Orders. Recharge charges by Shopify order id.
- [Stripe](https://maplith.com/integrations/md/stripe): From Recharge · Charges. Stripe, Braintree or other payment provider charges by transaction id.
- [HubSpot](https://maplith.com/docs/md/systems/hubspot): From Recharge · Customers. Email marketing and CRM contacts by customer email.
- [Pipedrive](https://maplith.com/docs/md/systems/pipedrive): From Recharge · Customers. Email marketing and CRM contacts by customer email.

Map Recharge with the rest of your business: https://maplith.com/integrations