Provider REST API

10 actions

Connect Claude to Infisical

Read, add, and rotate the secrets in a project's environments. 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: Access token
Actions exposed: 10
Cost per call (typical): 1 credit
Adapter: Maintained by Toolspoke

Connected in three steps

1
Install Infisical
Open the marketplace in your workspace, add Infisical to the project your agents work in, and it appears on the gateway immediately.
2
Connect the credential
Authenticate with access token. 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. Infisical 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 Infisical asks for

Access token. 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.

Access tokenRequired: A machine identity access token, or a service token. Grant it only the projects and environments this connection should reach - the token's own permissions are the real limit on every action here.; st.… or the token from a universal-auth login
Instance URLOptional: Leave blank for Infisical Cloud (US). Use https://eu.infisical.com for the EU region, or your own root URL for a self-hosted instance - without /api.; https://infisical.example.com
Default project IDOptional: Used when a call names no project. Find it with list_projects, or in the project's URL.; a1b2c3d4-…
Default environmentOptional: Environment slug used when a call names none, e.g. dev, staging, prod.; dev

What Claude can do in Infisical

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: 6Reads
Writes: 3Writes
Destructive: 1Destructive

Reads

Fetches data and changes nothing.

list_projects
List the projects this token can reach, each with its environments. Call this first - every secret action needs a project id and an environment slug, and the environment slugs come back here.
get_project
Fetch one project by id, with its environments and settings. Use list_projects when the id is not known yet.
list_secrets
List the secret keys in one environment and path. Values are withheld: this asks Infisical not to send them, and every result of this call is written to the workspace audit log. Read one value on purpose with get_secret instead, which is not logged. Set recursive to include every folder below the path.
get_secret
Read one secret's value by name. Use it only when the value is genuinely needed - this is the one action here that returns plaintext. The response is deliberately never written to the audit log, so the value exists only in this conversation. Set expand_references to resolve ${OTHER_SECRET} placeholders in the value.
list_folders
List the folders under one path in an environment. Folders are how a project groups secrets by service or component, and their names are the secret_path values every secret action takes.
list_audit_logs
Read the organisation's audit log - who read, changed, or deleted which secret, and when. Use it to answer "who touched this key" and to check that a rotation actually reached everything that reads it. Filters narrow it by project, environment, secret path or key, actor, and date range.

Writes

Creates or updates something on the other side.

create_secret
Add a new secret to one environment and path. Fails if the key already exists there - call update_secret for that. Anything deployed from this environment picks the value up on its next read.
update_secret
Change an existing secret's value, comment, or name. This is how a credential is rotated. The previous version is kept, so get_secret can still read it back by version - nothing is lost here.
create_folder
Create a folder in one environment, so a group of secrets can live under its own path. Creates nothing else - the secrets inside it are added with create_secret.

Destructive

Deletes or permanently alters something. Worth granting on purpose.

delete_secret
Delete a secret from one environment and path, with every version of its value. Whatever reads that key next gets nothing, which usually means an outage rather than an error anyone sees. There is no undo - rotate with update_secret unless the key genuinely has to stop existing.

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 Infisical 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 Infisical 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 Infisical: 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 Infisical. An agent has to ask.
It does not smooth over provider limits: Toolspoke does not retry, queue or back off around Infisical's own rate limits. A call that Infisical refuses comes back to the agent as a failed call.

Before you connect it

What can Claude do in Infisical?

10 named actions: 6 that only read, 3 that write and 1 that delete or permanently alter something. They include list_projects, get_project and list_secrets. 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 Infisical connector need?

Access token. The connector asks for access token, and optionally instance url, default project id and default environment. 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 Infisical 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 Infisical connector not do?

The 10 actions above are the whole of it. A call to any other name is refused before it reaches Infisical 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 Infisical 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 Infisical. An agent has to ask. Toolspoke does not retry, queue or back off around Infisical's own rate limits. A call that Infisical 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 Infisical'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 Infisical?

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 security connectors

Browse all of them