mcp2cli

mcp2cli is a Python CLI that turns an MCP server, an OpenAPI spec, or a GraphQL endpoint into executable CLI subcommands at runtime (no codegen). It can list tools/operations, call them with arguments, supports OAuth, caching (for specs/tool lists), “bake mode” to save connection configs, and produces token-efficient output modes for LLMs (e.g., TOON).

Evaluated Mar 30, 2026 (0d ago)
Homepage ↗ Repo ↗ DevTools cli mcp openapi graphql oauth agent-tools runtime-introspection caching
⚙ Agent Friendliness
62
/ 100
Can an agent use this?
🔒 Security
70
/ 100
Is it safe for agents?
⚡ Reliability
39
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
80
Documentation
75
Error Messages
0
Auth Simplicity
70
Rate Limits
10

🔒 Security

TLS Enforcement
80
Auth Strength
70
Scope Granularity
60
Dep. Hygiene
60
Secret Handling
75

Uses HTTPS URLs for remote endpoints by convention (TLS assumed via http client); supports OAuth flows and caching of tokens on disk. Secret handling includes env:/file: prefixes to avoid leaking secrets in process listings/CLI args. However, there is no explicit mention of secure token storage, redaction coverage, or rate limiting controls; MCP stdio mode runs local commands, increasing risk if inputs are untrusted.

⚡ Reliability

Uptime/SLA
0
Version Stability
70
Breaking Changes
40
Error Recovery
45
AF Security Reliability

Best When

You want an agent to dynamically discover and call tools across MCP/OpenAPI/GraphQL without generating client code, and you benefit from token-efficient outputs and cached introspection.

Avoid When

You need a strongly governed SDK/typed client with strict compile-time contracts, or you cannot risk local command execution (for MCP stdio mode) or cannot store OAuth tokens/caches on disk.

Use Cases

  • Give an AI coding agent a discoverable command-line interface for an MCP server
  • Rapidly expose an OpenAPI API as a CLI for automation and agent tool-calling
  • Interact with GraphQL queries/mutations from the command line with auto-generated selection sets
  • Cache introspection results (specs/tool lists) to reduce repeated discovery latency
  • Wrap frequently used API connections as baked CLI tools

Not For

  • Production API security gateway (it is a client-side CLI, not a policy enforcement layer)
  • High-integrity environments requiring strict guarantees about idempotency or transaction semantics (depends on target API/tool)
  • Usage scenarios where you cannot safely run arbitrary stdio MCP server commands (mcp-stdio involves executing a command)

Interface

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

Authentication

Methods: OAuth (authorization code + PKCE) OAuth client credentials Custom HTTP auth headers via --auth-header (supports env:/file: prefixes) MCP stdio server environment variables via --env
OAuth: Yes Scopes: Yes

OAuth tokens are persisted to ~/.cache/mcp2cli/oauth/ and refresh automatically; custom auth headers are supported. The README indicates secret support via env:/file: prefixes to reduce shell-argument exposure.

Pricing

Free tier: No
Requires CC: No

Open-source CLI (MIT). No hosted pricing described.

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • When using MCP stdio mode (--mcp-stdio), the CLI executes a local command to start the MCP server; agents should avoid untrusted commands/parameters.
  • Subcommands are generated dynamically from discovered tools/operations; agents should run --list (or --search) before assuming availability/argument shapes.
  • OAuth token persistence and cache locations are local; agents should be aware of filesystem permissions and multi-user environments.

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for mcp2cli.

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.

6447
Packages Evaluated
19956
Need Evaluation
586
Need Re-evaluation
Community Powered