Provider REST API

15 actions

Connect Claude to ConfigCat

Read and change feature flags, their per-environment values, configs and the audit log in ConfigCat. Toolspoke puts 15 of its actions behind one MCP endpoint that Claude, Cursor, and Codex all speak.

Connect View pricing

Connection: Provider REST API
Authentication: Management API credentials
Actions exposed: 15
What a call bills: Nothing
Adapter: Maintained by Toolspoke

Connected in three steps

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

Management API credentials. 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.

Basic auth username and passwordRequired: ConfigCat → your avatar → Public API credentials (https://app.configcat.com/my-account/public-api-credentials) → Add new credentials. Paste the Basic auth username, a colon, then the password. The credentials act with your own permissions in every organization you belong to, so a member of a read-only permission group gets read-only access here.; MDAwMDAwMDAtMDAwMC…:abc123…

What Claude can do in ConfigCat

15 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: 12Reads
Writes: 2Writes
Destructive: 1Destructive

Reads

Fetches data and changes nothing.

list_organizations
List the organizations these credentials belong to. The organization id is what audit and member reads at organization level take.
list_products
List every product these credentials can see, across all organizations. Start here - the product id is what configs, environments, tags, segments and audit logs are listed under - and it is the install's connection test, because it fails when the credentials reach nothing.
list_configs
List the configs in one product. A config holds a set of feature flags and settings; its id is what list_settings and get_config_values take.
list_environments
List the environments in one product, such as Test and Production. Every value read or change names one environment.
list_settings
List the feature flags and settings defined in one config, with each one's numeric id, key, type and tags. Values are per environment and are read separately.
get_setting
Read one feature flag or setting's definition - key, name, hint, type and tags - by its numeric id.
get_setting_value
Read what one flag or setting serves in one environment - its defaultValue, its ordered targetingRules and the percentage evaluation attribute. The rules are evaluated in the order returned.
get_config_values
Read the values of every flag and setting in one config for one environment, in a single call. Cheaper than calling get_setting_value once per flag.
list_tags
List the tags defined in one product. A tag id is what create_setting takes to label a flag.
list_segments
List the user segments defined in one product. Segments are reusable targeting conditions that targeting rules refer to by id.
list_stale_flags
List flags in one product that have not changed for a while - the candidates for removing from code and then deleting.
list_audit_logs
List the audit log for one product - who changed which flag, config or environment, and when - optionally narrowed to one config, one environment or a time window.

Writes

Creates or updates something on the other side.

update_setting_value
Change what one flag or setting serves in one environment, with a list of JSON Patch operations. Only defaultValue, targetingRules and percentageEvaluationAttribute can be changed. To switch a boolean flag on, send [{"op": "replace", "path": "/defaultValue/boolValue", "value": true}]; string, int and double settings use stringValue, intValue and doubleValue. This takes effect for every SDK reading the environment, so read the current value first with get_setting_value. A write, not destructive: the previous value can be set back, and the audit log records the change.
create_setting
Create a feature flag or setting in one config. It starts with the type's empty value in every environment and serves nothing new until a value is set, so creating one changes no running app.

Destructive

Deletes or permanently alters something. Worth granting on purpose.

delete_setting
Delete a feature flag or setting, with its values in every environment. Destructive because there is no undo, and any SDK still reading the key falls back to its own default from then on.

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 15 actions above are the whole of it. A call to any other name is refused before it reaches ConfigCat 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 ConfigCat 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 ConfigCat: 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 ConfigCat. An agent has to ask.
It does not smooth over provider limits: Toolspoke does not retry, queue or back off around ConfigCat's own rate limits. A call that ConfigCat refuses comes back to the agent as a failed call.

Before you connect it

What can Claude do in ConfigCat?

15 named actions: 12 that only read, 2 that write and 1 that delete or permanently alter something. They include list_organizations, list_products and list_configs. 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 ConfigCat connector need?

Management API credentials. The connector asks for basic auth username and password. 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 ConfigCat connector work with Cursor and Codex, or only Claude?

Any client that speaks MCP, and every one of them gets the same 15 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 ConfigCat connector not do?

The 15 actions above are the whole of it. A call to any other name is refused before it reaches ConfigCat 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 ConfigCat 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 ConfigCat. An agent has to ask. Toolspoke does not retry, queue or back off around ConfigCat's own rate limits. A call that ConfigCat 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 ConfigCat'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 ConfigCat?

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 what the call cost in sandbox time and model tokens. 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 developer tools connectors

Browse all of them