Provider REST API

10 actions

Connect Claude to SendGrid

Send email, and read deliverability stats, suppressions, templates and contacts. Toolspoke puts 10 of its actions behind one MCP endpoint that Claude, Cursor, and Codex all speak.

Connect View pricing

Connection: Provider REST API
Authentication: API key
Actions exposed: 10
Cost per call (typical): 1 credit
Adapter: Maintained by Toolspoke

Connected in three steps

1
Install SendGrid
Open the marketplace in your workspace, add SendGrid to the project your agents work in, and it appears on the gateway immediately.
2
Connect the credential
Authenticate with api key. Where to get one, and what it has to be able to reach, is the next section.
3
Point your agent at the gateway
Give your client one address, https://toolspoke.com/mcp. Claude Code takes it as a command, Claude and Claude Desktop add it as a custom connector, and Cursor, Codex and VS Code each read it from a config file of their own.

.mcp.json

{
  "mcpServers": {
    "toolspoke": {
      "type": "http",
      "url": "https://toolspoke.com/mcp"
    }
  }
}

One block covers every tool you have installed. SendGrid shows up in the client as soon as your policy allows it, and so does everything else you install later.

Where the address goes, per client

Claude Code: Run it in your project, then /mcp to sign in
claude mcp add --transport http toolspoke https://toolspoke.com/mcp
Claude and Claude Desktop: Settings, then Connectors, then Add custom connector
https://toolspoke.com/mcp
Cursor: ~/.cursor/mcp.json, or .cursor/mcp.json for one project
{ "mcpServers": { "toolspoke": { "url": "https://toolspoke.com/mcp" } } }
Codex: ~/.codex/config.toml
[mcp_servers.toolspoke] url = "https://toolspoke.com/mcp"
VS Code: .vscode/mcp.json, or the MCP: Add Server command
{ "servers": { "toolspoke": { "type": "http", "url": "https://toolspoke.com/mcp" } } }

What SendGrid asks for

API key. You provide it once, when you install the connector. Toolspoke encrypts it at rest and decrypts it only for the length of a single call, and the gateway attaches it to the outbound request itself, so it is never part of the arguments an agent sends.

API keyRequired: Settings → API Keys → Create API Key. Restricted Access is enough: Mail Send for send_mail, Suppressions for the bounce actions, Template Engine for the template reads, Marketing for the contact actions, and Stats for get_stats. The key is shown once.; SG.…

What Claude can do in SendGrid

10 actions, each one declared and named by the connector rather than discovered at runtime. A workspace policy grants a person all of them, a hand-picked selection, everything on the read side, everything on the write side, or none.

Reads: 8Reads
Writes: 0Writes
Destructive: 2Destructive

Reads

Fetches data and changes nothing.

list_bounces
List bounced addresses on this account with the reason and status the receiving server gave. This is the first thing to read when a customer says they never got the mail.
list_templates
List transactional templates with their ids and versions. Call it to find the template_id send_mail should render. SendGrid requires a page size on this endpoint, so one is always sent.
get_template
Retrieve one transactional template by id, with every version and the substitution variables each expects.
list_contacts
Return a sample of this account's marketing contacts - up to 50 of the most recent, sorted by email address. SendGrid takes no arguments here and offers no paging, so this is a sample rather than an enumeration; use search_contacts to find a specific person.
search_contacts
Find marketing contacts with a SendGrid query (SGQL), e.g. email LIKE '%acme.com%' or last_name = 'Wozniak'. Returns up to 50 contacts. A POST that reads and changes nothing, so it is classified read. SendGrid answers 408 if the query takes longer than 20 seconds.
get_stats
Return global email statistics for a date range: requests, delivered, opens, clicks, bounces, spam reports and unsubscribes. start_date is required by SendGrid; end_date defaults to today.
list_verified_senders
List the sender identities on this account, verified and unverified alike, each with the address and the verified flag. Call it before send_mail: an unverified from_email is the usual reason a send is refused.
get_account_details
Return the account's plan type and reputation score. Takes no arguments, reads nobody's mail and costs nothing, which is why it is what an install tests itself with.

Destructive

Deletes or permanently alters something. Worth granting on purpose.

send_mail
Send one email through SendGrid v3. Classified destructive rather than write, and deliberately so: it delivers to real inboxes, it cannot be recalled once SendGrid accepts it, and it consumes the account's send allowance. A write grant reaches nothing in this file. Supply html, text, or both - or a template_id that carries the body itself. SendGrid answers 202 with an empty body on success, so a successful send returns no message id.
delete_bounce
Remove one address from the bounce suppression list, so SendGrid will attempt delivery to it again. Destructive on both counts: the suppression record is gone with no undo, and lifting a suppression is what starts mail flowing to an address a receiving server already refused, which is how a sending reputation is damaged. Deletes exactly one address - this connector exposes no bulk or delete-all form. SendGrid answers 204 with an empty body.

What it will not do

Enforced by the gateway rather than left to convention, which is why each of these can be stated flatly.

It cannot call anything else: The 10 actions above are the whole of it. A call to any other name is refused before it reaches SendGrid rather than forwarded on, and connecting your account does not add to the list: it is fixed by the connector, not discovered at run time.
It reaches no further than your credential: Toolspoke holds no access to SendGrid of its own. Every call carries the credential you stored and nothing besides, so whatever that credential cannot reach, this connector cannot reach either.
It never hears from SendGrid: Nothing is pushed to it. There is no webhook, no subscription and no polling, so this connector cannot notice by itself that something changed in SendGrid. An agent has to ask.
It does not smooth over provider limits: Toolspoke does not retry, queue or back off around SendGrid's own rate limits. A call that SendGrid refuses comes back to the agent as a failed call.

Before you connect it

What can Claude do in SendGrid?

10 named actions: 8 that only read and 2 that delete or permanently alter something. They include send_mail, list_bounces and delete_bounce. Nothing outside that list is reachable: the connector declares each operation by name rather than proxying whatever an agent asks for.

What credentials does the SendGrid connector need?

API key. The connector asks for api key. Values are encrypted at rest and attached to the outbound request by the gateway, so they are never part of the arguments an agent sends and never reach the audit log.

Does the SendGrid connector work with Cursor and Codex, or only Claude?

Any client that speaks MCP, and every one of them gets the same 10 actions. There is a single address, https://toolspoke.com/mcp. Claude Code adds it with claude mcp add --transport http, Claude and Claude Desktop take it as a custom connector in settings, Cursor reads it from .cursor/mcp.json, Codex from ~/.codex/config.toml, and VS Code from .vscode/mcp.json. Each of them signs in to the gateway itself, so there is no key to paste.

What does the SendGrid connector not do?

The 10 actions above are the whole of it. A call to any other name is refused before it reaches SendGrid rather than forwarded on, and connecting your account does not add to the list: it is fixed by the connector, not discovered at run time. Toolspoke holds no access to SendGrid of its own. Every call carries the credential you stored and nothing besides, so whatever that credential cannot reach, this connector cannot reach either. Nothing is pushed to it. There is no webhook, no subscription and no polling, so this connector cannot notice by itself that something changed in SendGrid. An agent has to ask. Toolspoke does not retry, queue or back off around SendGrid's own rate limits. A call that SendGrid refuses comes back to the agent as a failed call.

Can I limit which actions an agent can call?

Yes, in two places. The project switches SendGrid's actions on and off one at a time, for everyone in the project at once, and the screen groups them by read, write and destructive so turning off everything that deletes is one click. An individual agent key can then be narrowed further, to particular toolkits in a project and to particular actions in a toolkit. Whatever it was granted, a key never reaches a project its owner cannot.

What gets recorded when an agent calls SendGrid?

Every attempt, with the agent that made it and the person that agent belongs to, the full request payload, the response payload, the status, the duration, and the credits spent. Values whose key names a secret are masked out before the record is shown to anyone. An operation the connector marks as not retained never has its response body written at all, so the gateway keeps no second copy of what was read.

More communication connectors

Browse all of them