SSO
World
Connect World once and hand every teammate a governed MCP endpoint with 31 tools ready to call from Claude, ChatGPT, Cursor, or any MCP client, inside the same permissions they already have, with every call checked and logged.
Automations & dashboards
From answering questions to doing the work
A person no longer has to ask. A trigger starts the work, inside the same permissions and the same audit log as everything else. Automations and live dashboards ship on the Black plan.
Automations
Put World work on a schedule.
A trigger starts a sequence of steps: call a World tool, transform the result, branch, wait, and ask a person when it matters.
World digest
Run 418 · started 2 minutes ago · on behalf of Emma Laurent
-
✓
Schedule
Every weekday at 08:00
0.2s -
✓
Fetch verifies
World
1.4s -
✓
Group by owner
Transform
0.1s -
✓
Draft the digest
Agent step
Ran with 4 tools, returned a structured summary
6.2s -
Approve the digest
Needs approval
Assigned to Michael Brennan
Approve Deny -
Post the digest
World
Queued
Collections and dashboards
Report on World without asking.
Collections hold your own data, written by an automation and read by a dashboard that refreshes on a schedule.
World health
Refreshed 4 minutes ago · every 15 minutes · from the verifies collection
Verifies
1,284 ↓ 12%
Verifieses
96 ↓ 8%
Needs attention
3 ↑ 2
Updated this week
412 ↑ 9%
Verifies created
Last 14 days
By workspace
Share of activity
Engineering 34%
Product 27%
Support 21%
Sales 18%
AI tools
World tools for your AI agents
31 tools are ready to call through Elaichi's MCP endpoint the moment you connect World, governed by the same roles, restrictions, and audit log as everything else in Elaichi.
create_a_world_verify
Verify a World ID proof for a Cloud action in World. Returns: success, action, nullifier_hash, created_at. Required: verify_id.
world_verifies_create_legacy
Verify a World ID proof for a Cloud action using the legacy v2 verify endpoint in World. Returns: success, action, nullifier_hash, created_at. Required: verify_id.
list_all_world_verifies
List World ID verifications in World. Returns records with: success, action, nullifier, created_at, environment, session_id, results, message.
create_a_world_minikit_send_notification
Send notifications to users of your World mini app by wallet address. Returns per-recipient results with walletAddress, sent, and reason. Required: app_id, wallet_addresses, mini_app_path. Either title and message (legacy) or localisations (localized) must also be provided; if both are given, localisations is used. Max 1000 wallet addresses per call.
get_single_world_minikit_transaction_by_id
Get a World MiniKit transaction by id to query its current on-chain status. Returns the transaction object including reference, transaction_hash, transaction_status, from, to, token_amount, token, chain, timestamp, and app_id. Required: id, app_id.
get_single_world_minikit_userop_by_id
Get a single MiniKit user operation in World by id and resolve the final on-chain transaction hash when it becomes available. Returns: status, userOpHash, sender, transaction_hash, nonce. Required: id.
list_all_world_transaction_debugs
Debug transactions that failed during the prepare stage in World. Returns a list of debug entries with debugUrl, createdAt, block, simulationRequestId, simulationError, and walletAddress. Required: app_id.
list_all_world_miniapps_prices
Get latest token prices in various fiat currencies from World. Returns: prices. Required: fiatCurrencies, cryptoCurrencies.
list_all_world_minikit_user_grant_cycles
Get the next grant claim cycle date for a user of your World mini app. Returns the user's humanity (orb-verified) grant cycle date if available, otherwise falls back to their document (passport-verified) grant cycle date. Returns: nextGrantClaimUTCDate. Required: wallet_address, app_id. The wallet_address must be exactly 42 characters long.
create_a_world_create_action
Create an incognito action for an app in World. Returns the created action object with its id and content-type-specific attributes. Required: create_action_id.
create_a_world_precheck
Fetch action metadata for a World ID app to determine a user's eligibility for verification. Returns: id, is_staging, is_verified, name, verified_app_logo, engine, __typename, sign_in_with_world_id, can_user_verify, action. Required: precheck_id, action.
list_all_world_jwks
List JWK public keys used to verify World JWT signatures for verification requests. Returns: kid, kty, e, n.
create_a_world_graphql
Execute GraphQL queries and mutations against the World Developer Portal via the GraphQL proxy. Returns: data (query results whose shape depends on the operation), errors (array of GraphQL error objects if any occurred). Required: query.
update_a_world_cancel_recovery_agent_update_by_id
Cancel a pending recovery agent update in the World ID gateway. Returns: request_id, kind, status. Requires a JSON request body conforming to the CancelRecoveryAgentUpdateRequest schema.
create_a_world_create_account
Create a new World ID account via the World registry gateway. Returns a 202 gateway status response including request_id, kind, and status for tracking the async on-chain request. Required: recovery_address, authenticator_addresses, authenticator_pubkeys, offchain_signer_commitment. Numeric string fields accept decimal or 0x/0X-prefixed hex.
update_a_world_execute_recovery_agent_update_by_id
Execute a pending recovery agent update for a World ID account in the World gateway. Returns: request_id, kind, status (with state and optional tx_hash). Required: leaf_index, old_offchain_signer_commitment, new_offchain_signer_commitment, signature, nonce. Returns 400 if leaf_index is zero or no pending update exists.
list_all_world_healths
Check the health status of the World gateway server. Returns a general server status check response indicating server availability.
update_a_world_initiate_recovery_agent_update_by_id
Initiate a recovery agent update for a World ID account through the World gateway. Returns a gateway status object including request_id, kind, and status for polling the on-chain request. A JSON request body (UpdateRecoveryAgentRequest) is required.
create_a_world_insert_authenticator
Insert a new authenticator for a World account via the registry gateway. Returns a 202 status response with request_id, kind, and status. Required: leaf_index, new_authenticator_address, old_offchain_signer_commitment, new_offchain_signer_commitment, signature, nonce, pubkey_id, new_authenticator_pubkey. Numeric string fields accept decimal or 0x-prefixed hex.
list_all_world_is_valid_roots
Check whether a given merkle root is currently valid on-chain in World. Returns: valid. Required: root.
create_a_world_recover_account
Recover a World account by submitting a recovery request to the registry gateway. Returns: request_id, kind, status. Required: leaf_index, new_authenticator_address, old_offchain_signer_commitment, new_offchain_signer_commitment, signature, nonce. The request is processed asynchronously and returns 202 Accepted.
create_a_world_remove_authenticator
Remove an authenticator from a World ID account via the World gateway. Returns: request_id, kind, status. Required: leaf_index, authenticator_address, old_offchain_signer_commitment, new_offchain_signer_commitment, signature, nonce. Numeric string fields accept decimal or 0x/0X-prefixed hex; returns 202 Accepted with a request_id for polling status.
get_single_world_status_by_id
Get the status of a World registry gateway request by id. Returns the request_id, kind (the operation type), and status (current state with optional tx_hash and error details). Required: id.
create_a_world_update_authenticator
Update an authenticator in the World ID registry gateway. Returns: request_id, kind, status. Required: leaf_index, old_authenticator_address, new_authenticator_address, old_offchain_signer_commitment, new_offchain_signer_commitment, signature, nonce, pubkey_id, new_authenticator_pubkey. Numeric string fields accept decimal or 0x-prefixed hex. Response is asynchronous (HTTP 202) — poll request_id…
create_a_world_authenticator_pubkey
Get the compressed authenticator public keys for a World ID account by leaf index from the World ID Indexer. Returns: authenticator_pubkeys, offchain_signer_commitment. Required: leaf_index.
create_a_world_inclusion_proof
Get a Merkle inclusion proof for a World ID account by leaf index from the current WorldIDRegistry tree, including the full authenticator public key slot list with null entries for removed authenticators. Returns: authenticator_pubkeys, leaf_index, root, siblings. Required: leaf_index.
create_a_world_packed_account
Get packed account data for a given authenticator address from the WorldIDRegistry contract in World. Returns: packed_account_data. Required: authenticator_address.
create_a_world_pending_recovery_agent
Get the pending recovery agent update for a particular World ID given its leaf index in World. Returns: execute_after, pending_recovery_agent, recovery_agent_address. If no recovery agent update is pending, the zero address and zero execute-after timestamp are returned. Required: leaf_index.
create_a_world_recovery_agent
Get the recovery agent address for a particular World ID given its leaf index from the World ID Indexer. Returns: recovery_agent. Required: leaf_index.
create_a_world_signature_nonce
Get the current signature nonce for a World ID by its leaf index from the World ID Indexer. The nonce is used to perform on-chain operations for the World ID. Returns: signature_nonce. Required: leaf_index.
get_single_world_credit_by_id
Get credit data for a World borrower by id (wallet address or World username). Returns: state, score. Required: id.
No tools match your search.
How it works
From World account to governed MCP endpoint
Connect
Link a World account through Elaichi's hosted connect flow. Credentials are vaulted, so nobody, including the AI, ever sees them.
Compose
The connection becomes a toolbox instantly. Curate which World tools are exposed, rename them, or freeze arguments.
Authorize
Point Claude, ChatGPT, Cursor, or any MCP client at the one Elaichi endpoint and sign in with OAuth, governed by roles, restrictions, and audit logs.
Put your agents to work in the tools you already use
Start a 14-day Gold trial. Connect a product, curate a toolbox, and paste a governed MCP endpoint into Claude, ChatGPT, or Cursor in minutes.