How to connect Resend 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 Resend from the dashboard, then start a fresh chat if the runtime still has the old tool catalog loaded.

Resend 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 Resend 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.

Resend's own knowledge base names the causes, and three of them are account state rather than connection state. The first is key scope: a key restricted to read-only or to a specific project fails sending calls, and the fix is a full-access key. The second is domain verification: sending from a domain that is not verified in the account fails, and the fix is the domain's DNS records in Resend. The third is a from mismatch: the sender address must sit on a verified domain, and a from address on an unverified or foreign domain is refused. The fourth is the test sender: onboarding@resend.dev only works for testing and is the cause engines teach for 403s on first sends. The page's own cause completes the list: a placeholder argument such as YOUR_DOMAIN reaching the API literally. Re-paste the current key, check the domain and from address, then re-run before touching the connection.

Ask the agent to diagnose it:

Call resend_list_domains and tell me the verification status of each domain, then quote the exact error text from the failed call including the from address it used. Do not retry it yet.

Resend rate-limits sending and API traffic, and an agent that retries a failed send in a tight loop is the shape that meets the limit fastest. The response is to stop and look at the send itself first: failed sends are usually the account causes above (domain, from address, scope) rather than rate, and retrying the same send produces the same error. If the error is a genuine rate limit, the agent should report the retry detail and pace the work, batching where the tools allow, rather than hammering the endpoint.

Ask the agent to diagnose it:

Tell me exactly which Resend tools you called in the last few minutes and what the send or list pattern was. Quote the full error text from the last response. Do not call Resend again until I reply.

Two failures share this symptom. If no Resend tools appear at all, the connection or pairing is incomplete: confirm the plugin was installed with --enable, that hermes clawlink begin and hermes clawlink finish both ran, and that Resend shows as connected in the dashboard. If most tools work and one name fails, the name is wrong rather than missing, and the error lists the closest real ones. There is also a first-call timing case: schemas load on demand, so the opening Resend call in a fresh session can arrive before the catalog and needs one retry. The fixes search results teach for this symptom, reloading the MCP client and setting RESEND_API_KEY in an environment, belong to a local setup and do not apply here.

Ask the agent to diagnose it:

List the Resend tools you actually have access to. If there are none, say so plainly and tell me whether Resend shows as connected. Do not guess tool names.

Double-check that the API key for Resend has the right scopes or account access. A valid key can still be too limited for some reads or writes.

Yes. ClawLink is the fastest way to connect Hermes to Resend: link your Resend account once in the browser and Hermes Agent can call the Resend API through 62 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 Resend in the dashboard and Hermes can use it from the next message — no config files, and the Resend key you paste is stored server-side instead of in your environment.

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

The alternative to ClawLink is usually manual API key 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 Resend working from chat.

Resend's hosted MCP is a real route: a streamable HTTP endpoint, browser OAuth, no raw key handled by the user. The difference is the surface and the management. Their MCP is one endpoint registered in a client config; this page's route exposes 62 named tools Hermes can reason about individually, and it puts Resend beside your other connected accounts with one dashboard to manage and revoke. If you run a plain MCP client and want Resend there verbatim, their hosted server is the direct choice. If Hermes is your agent and you want Resend alongside everything else, this is the shape that fits.

A Resend API key from the Resend dashboard, pasted once on the hosted setup page and held server-side. It is the same key the manual routes put in a config file or environment; the difference is that it never sits on a machine. Rotating it in Resend means updating one place, and disconnecting from the dashboard revokes the agent's use of it immediately.

Send email, manage domains and their verification, create and manage audiences, contacts, and templates, read sent and received email lists and attachments, manage webhooks, and read API keys. The sending half needs two account prerequisites that are the usual 403 causes: a verified domain and a from address on that domain. The management half is where prompts deserve care, because creating domains, audiences, and templates is account state that other work depends on.

The tool set includes resend_list_received_emails and resend_get_email_attachment, so received messages are readable when the account has inbound enabled. That is not the same as full inbox semantics with reply threading, which belongs to the agent's email reading side; for sending plus inbound visibility, this integration covers both ends of the loop.

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.