npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@call-e/cli

v0.5.2

Published

CLI wrapper for brokered MCP login and CALL-E MCP client configuration.

Readme

@call-e/cli

@call-e/cli ships the calle command.

The CLI is not an OAuth server and not an MCP server. It is a local wrapper that uses the CALL-E broker API to complete browser-based login, cache the OAuth token locally, print MCP client configuration, and provide call workflow shortcuts for terminal-based agents.

For install and authentication steps, see docs/install/cli.md.

Quick Start

First follow CLI entry point selection to prepare the launcher and request.json. Use each array below as the request's argv, then run node run-agent-command.mjs request.json to authenticate, inspect parameters, or plan a call:

["auth", "login"]
["call", "plan", "--help"]
["call", "plan", "--to-phone", "+15551234567", "--goal", "Confirm the appointment"]

Help is available at every command level:

["--help"]
["call", "--help"]
["call", "plan", "--help"]

Commands

["--version"]
["auth", "login"]
["auth", "login", "--start-only", "--no-browser-open"]
["auth", "status"]
["auth", "logout"]
["mcp", "config"]
["mcp", "tools"]
["mcp", "call", "plan_call", "--args-json", "{\"to_phones\":[\"+15551234567\"],\"goal\":\"Confirm the appointment\"}"]
["call", "plan", "--help"]
["call", "plan", "--to-phone", "+15551234567", "--goal", "Confirm the appointment"]
["call", "start", "--to-phone", "+15551234567", "--goal", "Confirm the appointment"]
["call", "run", "--plan-id", "<plan_id>", "--confirm-token", "<confirm_token>"]
["call", "recover", "--recovery-id", "<recovery_id>"]
["call", "status", "--run-id", "<run_id>"]
["regions", "list"]

Defaults:

  • Base URL: https://seleven-mcp-sg.airudder.com
  • MCP channel: openagent_oauth
  • MCP server URL: <baseUrl>/mcp/openagent_oauth
  • Broker API: <baseUrl>/api/v1/openagent-auth/*
  • Token cache: ~/.calle-mcp/cli

calle auth login opens the brokered login URL, polls the broker session, exchanges the authorized session, and stores the token in a private local cache. The token is not printed to stdout.

calle auth login --start-only --no-browser-open creates or reuses a brokered login session and returns JSON with login_url and assistant_hint.message without polling for completion. This is intended for agent integrations that need to show the authorization link before continuing.

calle mcp config prints a JSON MCP client config:

["mcp", "config", "--base-url", "https://seleven-mcp-sg.airudder.com"]

Example output:

{
  "mcpServers": {
    "calle": {
      "type": "http",
      "url": "https://seleven-mcp-sg.airudder.com/mcp/openagent_oauth"
    }
  }
}

For LLM clients that can connect to MCP directly, prefer calle mcp config and let the client use the MCP tool schemas. For terminal-based agents such as Codex, the calle call ... commands provide shortcuts over the same remote MCP tools.

For agent-facing outbound calls, prefer calle call start. It performs planning and execution inside one CLI invocation and does not print execution confirmation data.

If execution may have been accepted but no run_id was received, the CLI returns retry_safe: false with an opaque recovery_id and next_argv. Use that array as the next request's argv instead of repeating call start; it reuses the original confirmation context without printing it. If only the initial status query fails, the command still returns the accepted run_id and a call status next_argv array. Use the same launcher for that status request.

Successful command stdout is JSON except help and version output. Some top-level or local failures may print plain stderr. Access tokens are read from the local cache and are never printed.

Options

See docs/cli-reference.md for the canonical command and option reference, including defaults, required arguments, advanced configuration, and per-command examples.

Telemetry / Usage Data

The CLI sends best-effort usage telemetry to the configured CALL-E service at <base-url>/api/ui-telemetry/track to help diagnose installation, authentication, MCP tool availability, and drop-off before a first plan_call reaches the server.

Collected fields include an anonymous installation ID stored under the CLI cache root, CLI version, integration source, command stage, outcome, error type, and the endpoint metadata below. The payload does not include phone numbers, call goals, OAuth tokens, broker login URLs, full argument JSON, transcripts, or contact data.

| Field | Value | | --- | --- | | base_url_host | Hostname and any non-default port from the configured base URL, using new URL(value).host. | | server_host | Hostname and any non-default port from the configured server URL, using new URL(value).host. | | server_url_hash | SHA-256 of the complete configured server URL string, including any path, query, and fragment. |

For example, with base URL https://api.example.com:8443 and server URL https://mcp.example.com:9443/mcp?mode=test, the readable fields are api.example.com:8443 and mcp.example.com:9443. The hash is computed from the entire https://mcp.example.com:9443/mcp?mode=test string, not just its host. Default ports are omitted by URL parsing (for example, https://api.example.com:443 produces api.example.com). Hashing the server URL does not conceal the two separate readable host fields.

Disable CLI telemetry with DO_NOT_TRACK=1, CALLE_TELEMETRY=0, or --no-telemetry:

["auth", "status", "--no-telemetry"]
["mcp", "tools", "--no-telemetry"]

Broker and MCP requests still create service-side security, audit, and business operation logs needed to authenticate users and run calls.

Development

pnpm --filter @call-e/cli test
pnpm --filter @call-e/cli check
pnpm --filter @call-e/cli pack:dry-run

For offline smoke checks and live OAuth/MCP validation, see docs/cli-verification.md.