mcpshim

MCPShim provides a local daemon (mcpshimd) and CLI (mcpshim) that turn remote MCP server tools into standard command-line workflows. It manages MCP session/discovery, optionally OAuth login, and exposes tool calls via a JSON-over-UNIX-socket IPC protocol and mapped CLI flags, returning structured JSON output and maintaining local call history in SQLite.

Evaluated Mar 30, 2026 (22d ago)
Homepage ↗ Repo ↗ DevTools mcp ai-agents cli ipc oauth sqlite tool-calling
⚙ Agent Friendliness
58
/ 100
Can an agent use this?
🔒 Security
53
/ 100
Is it safe for agents?
⚡ Reliability
24
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
70
Documentation
75
Error Messages
0
Auth Simplicity
60
Rate Limits
20

🔒 Security

TLS Enforcement
60
Auth Strength
65
Scope Granularity
30
Dep. Hygiene
50
Secret Handling
55

Uses OAuth and configurable headers; tokens appear stored in local SQLite. The README does not describe secure storage controls (encryption at rest, filesystem permissions), token redaction in logs, or scope granularity. No explicit rate limiting guidance is given.

⚡ Reliability

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

Best When

You want a language-agnostic way (shell/CLI) for agents/scripts to call many different MCP servers with consistent auth and JSON I/O on a single machine.

Avoid When

You need direct network APIs, fine-grained cloud identity integration, or strict guarantees about idempotency/retries at the API-operation level without checking the daemon’s implementation.

Use Cases

  • Let LLM agents and scripts call remote MCP tools using simple CLI invocations
  • Centralize and standardize auth/session handling across multiple MCP servers
  • Reduce agent prompt/context overhead by using alias-based local command workflows instead of dumping raw MCP schemas
  • Persist and inspect tool call history locally for debugging/auditing

Not For

  • Projects needing a hosted SaaS/API surface (this is a local daemon + CLI pattern)
  • Environments where Unix socket IPC or local daemon processes are not allowed
  • Use cases requiring standardized REST/GraphQL/SDK integrations out of the box

Interface

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

Authentication

Methods: Configured request headers per server (e.g., Authorization header) OAuth login flow for OAuth-capable MCP servers (auto-initiated on 401 when configured appropriately) Manual/cross-device OAuth completion
OAuth: Yes Scopes: No

Auth is configured via CLI headers per server and tokens are stored locally in SQLite (oauth_tokens per README). Scope granularity is not described.

Pricing

Free tier: No
Requires CC: No

No pricing information provided; appears to be open-source/local software.

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • Relies on a local daemon + Unix socket; agent environments must support running/connecting to mcpshimd
  • Retry behavior is mentioned at the lifecycle level, but idempotency semantics for repeated tool calls are not documented in the provided README
  • JSON output behavior depends on whether stdout is a terminal; agents may need to use --json for consistent parsing

Alternatives

Full Evaluation Report

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

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