About Jeero MCP
Jeero MCP is a local MCP server for managing Jeero subscriptions and event data. It syncs with Mother and caches processed event records in SQLite for users who need to inspect and refresh event information from an MCP client.
Key features
- 01Manage multiple subscriptions
- 02Sync event data with Mother
- 03Cache event records in SQLite
- 04View events and ticket snapshots
- 05Configure credentials in a local browser
- 06Redact credentials from tool outputs
Use cases
- 01Inspect event information from an MCP client
- 02Refresh subscription data on demand
- 03Manage multiple event-data subscriptions
- 04Configure connection credentials locally
MCP support
Communicates with MCP clients over stdio and provides tools for subscription management, synchronization, event queries, ticket snapshots, and logs.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEJeero MCP
Jeero MCP is a local Model Context Protocol server for Jeero subscriptions and event data. It synchronizes with Mother and keeps its local event cache in SQLite; it does not require WordPress or a hosted MCP service.
Install
Use Node.js 20 or later and configure your MCP client to run the published package:
{
"mcpServers": {
"jeero": {
"command": "npx",
"args": ["-y", "jeero-mcp"]
}
}
}
The server communicates over stdio. Do not write application logs to stdout.
First use
Create a subscription with create_subscription, then call config_subscription without settings. Jeero opens a private configuration flow using MCP URL elicitation when the client advertises support. Otherwise the tool returns a link and an access code. Open the link yourself, enter the code, choose your ticketing solution, and save to load its connector-specific fields. Enter keys only in that browser page. Keep the page private: do not ask your agent or its browser tools to inspect or fill it.
The screen runs on 127.0.0.1 on the same computer as the MCP server. Links expire after 15 minutes or when the server stops. If an app's URL prompt does not work, call config_subscription with manual: true to get a fresh link. This local screen requires a browser on the MCP server's computer; it is not a remotely hosted configuration portal.
Only settings.theater can still be supplied through create_subscription or config_subscription. Other settings, including all credentials, must be entered in the browser. Existing saved settings continue to work. This intentionally replaces the previous arbitrary settings input and raw subscription envelopes: get_subscription and create_subscription now return a summary with configured field names, required-field completeness, and a connection-error count. Completeness does not prove that a key is valid; check synchronization too. Raw Mother metadata, field values, and saved credentials are never returned by these tools.
One subscription can be the default; tools use it when subscription is omitted. Multiple active subscriptions are supported.
get_events synchronizes the selected subscriptions at most once a minute. While the server is running, active subscriptions are also checked every minute. Use sync_subscription or sync_subscriptions with force: true when an immediate refresh is needed.
Tools
- Subscription management:
list_subscriptions,create_subscription,get_subscription,config_subscription,set_default_subscription,activate_subscription,deactivate_subscription - Synchronization:
sync_subscription,sync_subscriptions - Event data:
get_events,get_event,list_event_categories,get_ticket_snapshots,get_logs
Local data and configuration
By default Jeero MCP stores its SQLite cache and persistent site identity in ~/.jeero-mcp/. Keep that directory private: the identity is used for Mother requests.
Credentials remain in the local SQLite settings and are sent to Mother using its existing API. This change keeps them out of MCP tool answers and errors; it does not encrypt the database or erase keys from old chat histories or existing debug files. Tool outputs and newly imported logs redact known credentials; request tracing omits query strings and response bodies. No third-party scripts or assets are loaded by the configuration screen.
New installations use a random UUID as their site identifier, saved in site-identity.json and reused on every restart. Existing installations keep their saved identifier and key so their Mother subscriptions and inbox remain accessible. Do not copy this identity file between independent installations.
| Variable | Purpose |
|---|---|
JEERO_DATA_DIR | Override the local data directory. |
JEERO_SITE_URL | Optional site identifier override for new installations; defaults to a random UUID. Existing saved identities take precedence. |
JEERO_MOTHER_BASE_URL | Override Mother’s base URL, for example for staging. |
JEERO_NO_OF_ITEMS_PER_PICKUP | Limit inbox items requested per sync. |
JEERO_LOG_MOTHER | Enable Mother request tracing with true, 1, yes, or on. |
JEERO_LOG_MOTHER_FILE | Override the trace log path. |
Mother remains the upstream source of truth. Jeero MCP stores processed event records and synchronization state, not raw inbox payloads.
Development
npm install
npm test
npm pack --dry-run
The tag-based publish workflow requires an explicit repository release approval before a version tag can publish.
License
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.