How to connect Gumroad to Hermes Agent

Pairing is a two-step handshake: run hermes clawlink begin, approve the link in your browser, then run hermes clawlink finish. If you ran finish before approving, or the approval link expired, run hermes clawlink begin again to get a fresh link. Confirm the plugin was installed with --enable, then verify with hermes clawlink test.

Reconnect Gumroad from the dashboard, then start a fresh chat if the runtime still has the old tool catalog loaded.

Gumroad tool schemas load on demand the first time a tool runs and are cached after that, so this error usually clears on its own: wait a few seconds and retry the same request. If every Gumroad call keeps failing with it in a fresh chat, reconnect from the dashboard, and contact support if it still persists — that pattern points to a configuration problem on our side, not something you can fix by reconnecting again.

Two usual causes. The connected account may not have access to the specific workspace, inbox, store, or project in the request — check that first. If access looks right, the agent may have sent a placeholder value (like "YOUR_ID" or an example id from documentation) instead of a real one: ask it to run a list or search tool first, then retry the action with a real id from those results. Most failures at this stage are one of these two, not ClawLink bugs.

The taught fix set and this page's cause barely overlap, and both are real. The taught set: generate a fresh access token, verify the OAuth scopes, and disconnect and reconnect the third-party app, which matters when a grant predates a permission change. This page's cause, which no engine covers, is the placeholder argument: an invented or copied-from-an-example product, sale, or license id returns a permission-shaped error because Gumroad hides what the account cannot see. When a 403 appears with a healthy connection, check the id the agent sent before touching the connection, then the scopes, then reconnect.

Ask the agent to diagnose it:

Call gumroad_get_user and gumroad_list_products. Tell me which Gumroad account this connection acts as and how many products it can see, then quote the exact error from the failed call. Do not retry it yet.

Gumroad publishes no numeric rate limit, and it throttles with 429 when callers exceed its tolerance, so there is no number to plan around, only the behavior. An agent paging through sales is the shape that meets it, especially when it runs a verification loop over many license keys. The fix is pacing: spread pages across turns, have the agent report the page it is on, and treat a 429 as a stop-and-wait signal rather than a retry trigger.

Ask the agent to diagnose it:

Tell me how many Gumroad calls you have made recently and what you were reading (sales pages, license checks, or product lists). Quote any 429 text from the last response. Do not call Gumroad again until I reply.

Two different failures. If Hermes shows no Gumroad tools at all, the connection or the pairing is incomplete: confirm the plugin was installed with --enable, that both hermes clawlink begin and hermes clawlink finish ran, and that Gumroad shows as connected in the dashboard, then start a fresh chat. If most tools work and a single name fails, the name is wrong rather than missing, and the error lists the closest real ones. The search-result fix for this symptom, setting a GUMROAD_ACCESS_TOKEN env var or editing an mcp.json config, is the self-hosted version of the failure and does not apply: schemas here load on demand, so the opening Gumroad call in a fresh session can arrive before the catalog and needs one retry, then reconnect.

Ask the agent to diagnose it:

List the Gumroad tools you actually have access to. If there are none, say so plainly. If there are, tell me which one lists products and use that exact name.

Try reconnecting Gumroad and complete the consent flow in the same browser session. Partial OAuth approvals or switching accounts mid-flow can leave the connection incomplete.

Yes. ClawLink is the fastest way to connect Hermes to Gumroad: link your Gumroad account once in the browser and Hermes Agent can call the Gumroad API through 7 ready-made tools — no custom code or token handling.

Install the plugin with hermes plugins install ClawLink-HQ/hermes-plugin --enable, then pair once: run hermes clawlink begin, approve the link in your browser, and run hermes clawlink finish. Connect Gumroad in the dashboard and Hermes can use it from the next message — no config files or API keys to manage.

About two minutes. Sign in, click Connect next to Gumroad in the dashboard, authenticate, and Hermes Agent can use it from the next chat message.

The alternative to ClawLink is usually manual OAuth app setup plus your own token handling, permission troubleshooting, and tool plumbing for Hermes Agent. That is fine if you want to build and maintain the integration yourself. Most teams just want Gumroad working from chat.

No, and the token ritual is exactly what this page removes. The 8-step flow search results teach, create an application in Advanced Settings, generate a token, copy it once, is the self-hosted route's prerequisite. Gumroad's API also supports OAuth, and the hosted connection uses it: you approve the sign-in in the browser and any token handling happens server-side, out of sight and out of your env files. If you are comparing native options, Gumroad now also ships its own command-line tool with a login flow, which is the provider's own answer to the same question; this integration is the route that works from Hermes chat alongside your other connected apps.

The creator surface: list products, read the user record, retrieve sales, verify licenses, and manage resource subscriptions (subscribe, unsubscribe, and read existing subscriptions). There is no tool for editing product listings or payouts; the toolkit is deliberately a read-and-verify surface. License verification is the quiet power move: gumroad_verify_license answers "is this purchase valid" for a license key, which is the support-adjacent question sellers actually get.

It can read sales, with one provider fact worth knowing: Gumroad's sales endpoint returns 10 results per request, so a single gumroad_get_sales call is a page, not the whole ledger. An agent asked for "all sales" should page through repeatedly or say what the first page contains. There is no published numeric rate limit on Gumroad's API; it throttles with a 429 when a caller exceeds its tolerance, so pacing pages rather than hammering them is the reliable pattern.

The market's safety answer is "moderately safe with strict permissions," and the mechanics of this connection deliver the strictness: the tool table on this page is the complete set of operations the agent can invoke, revenue data is read through your connected account rather than a token that could leak from a config file, and revoking from the dashboard cuts access immediately. Because the toolkit is read-and-verify heavy, an agent connected here can summarize sales and check licenses without holding any write capability at all.

Yes, with the honest caveat that every AI agent touching Gumroad uses the API under the hood; what this page removes is the key, not the API. Search results answer that query with Gumroad's own first-party CLI and native tools, which are Gumroad's in-house options. This integration is the external-agent route: no key is created by you, the OAuth belongs to the hosted app, and the same tools work for Hermes as for any other agent client.

Pairing is a two-step handshake: run hermes clawlink begin, approve the link in your browser, then run hermes clawlink finish. If you ran finish before approving, or the approval link expired, run hermes clawlink begin again to get a fresh link. Confirm the plugin was installed with --enable, then verify with hermes clawlink test.