mcp-cli

mcp-cli is a Bun/TypeScript command-line tool that discovers and invokes Model Context Protocol (MCP) server tools. It supports MCP servers over stdio or HTTP, loads tool schemas on-demand, provides shell-friendly JSON output, and can use a per-server daemon/connection pooling via Unix sockets.

Evaluated Mar 30, 2026 (22d ago)
Repo ↗ DevTools mcp cli bun typescript tooling agent-optimized json stdio http
⚙ Agent Friendliness
72
/ 100
Can an agent use this?
🔒 Security
62
/ 100
Is it safe for agents?
⚡ Reliability
39
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
78
Documentation
86
Error Messages
--
Auth Simplicity
90
Rate Limits
20

🔒 Security

TLS Enforcement
80
Auth Strength
55
Scope Granularity
45
Dep. Hygiene
60
Secret Handling
70

mcp-cli supports injecting Authorization headers via config and environment-variable substitution, which can be secure if users avoid logging. However, the excerpt does not describe TLS enforcement details (e.g., cert validation) beyond using HTTPS URLs in examples, nor does it detail how secrets are masked in logs. Tool filtering can reduce blast radius but is not a substitute for server-side authorization.

⚡ Reliability

Uptime/SLA
0
Version Stability
60
Breaking Changes
40
Error Recovery
55
AF Security Reliability

Best When

You want lightweight, scriptable access to MCP tools from the command line (including agent workflows) and can manage MCP server configuration locally.

Avoid When

You need standardized HTTP endpoints, webhook delivery, or strong platform-level authz enforced by mcp-cli itself rather than by the underlying MCP servers.

Use Cases

  • Discover MCP servers and list available tool schemas
  • Inspect tool input schemas (info) before invocation
  • Invoke MCP tools from shell scripts and CI using JSON in/out
  • Chain multiple tool calls using pipes and jq
  • Filter/allow/disable tools per MCP server via configuration
  • Use with AI coding agents that can run shell commands

Not For

  • Building a long-running MCP gateway service (it’s a CLI, not a server)
  • Providing a direct REST/GraphQL API to your application users
  • Use as a security boundary for untrusted MCP servers (it orchestrates them)
  • Environments that require strict network egress controls without reviewing configured remote servers

Interface

REST API
No
GraphQL
No
gRPC
No
MCP Server
Yes
SDK
Yes
Webhooks
No

Authentication

Methods: Config-provided HTTP headers (e.g., Authorization: Bearer ...)
OAuth: No Scopes: No

Authentication is delegated to MCP server configuration (e.g., headers for HTTP transports). No first-class OAuth flow is provided by mcp-cli itself.

Pricing

Free tier: No
Requires CC: No

No pricing model described (CLI tool).

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Documented

Known Gotchas

  • Remember correct subcommands (use 'call' and 'info'; 'server tool' or 'run' are not valid)
  • When passing complex JSON, prefer stdin/heredoc to avoid shell-escaping issues
  • Tool filtering is applied globally—an agent may not see tools that are disabled/filtered in config
  • Remote HTTP servers require correct header configuration; missing env vars can fail unless MCP_STRICT_ENV=false

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for mcp-cli.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-03-30.

8642
Packages Evaluated
17761
Need Evaluation
586
Need Re-evaluation
Community Powered