mcp-server-openapi

Provides an MCP server that reads an OpenAPI 3.0/3.1 specification (JSON) and dynamically generates MCP tools for each API endpoint/method. When called by an MCP client (e.g., Claude Desktop), it validates parameters, constructs the HTTP request (including path/query/header/body), applies an optional configured Authorization header, and returns formatted JSON responses with errors.

Evaluated Apr 04, 2026 (16d ago)
Repo ↗ API Gateway mcp openapi tool-generation rest-api claude-desktop go
⚙ Agent Friendliness
61
/ 100
Can an agent use this?
🔒 Security
53
/ 100
Is it safe for agents?
⚡ Reliability
22
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
78
Documentation
70
Error Messages
--
Auth Simplicity
85
Rate Limits
10

🔒 Security

TLS Enforcement
80
Auth Strength
55
Scope Granularity
20
Dep. Hygiene
50
Secret Handling
60

README claims a 30-second timeout and 'Detailed error messages without exposing sensitive data' plus parameter validation and URL/header escaping. However, there is no provided evidence for TLS enforcement details, secure secret storage/logging behavior, dependency audit status, or defense against SSRF/path traversal via OpenAPI-defined URLs.

⚡ Reliability

Uptime/SLA
0
Version Stability
35
Breaking Changes
20
Error Recovery
35
AF Security Reliability

Best When

You have a reasonably complete OpenAPI JSON spec for a REST API and can authenticate using a static header value (Bearer/API key/custom header) configured at startup.

Avoid When

Your OpenAPI specs are incomplete/incorrect (e.g., missing operationId, parameter schemas) or you need OAuth2 flows, pagination semantics, retries/idempotency guarantees, and comprehensive rate-limit handling documented by the server.

Use Cases

  • Expose any REST API described by an OpenAPI spec as MCP tools for an LLM assistant
  • Rapidly integrate internal or third-party REST APIs into Claude Desktop via MCP
  • Generate tool schemas and argument validation automatically from OpenAPI
  • Provide a generic “API caller” interface for agents without writing bespoke tool wrappers

Not For

  • APIs that require complex auth flows not representable as a single static header/token
  • Non-REST or non-OpenAPI-described functionality (e.g., custom RPC protocols, streaming-only APIs)
  • Production deployments where you need a fully specified, reviewed MCP server contract, test coverage evidence, and explicit operational SLAs

Interface

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

Authentication

Methods: Static Authorization header (configurable via -auth) for Bearer tokens, API keys, or custom headers
OAuth: No Scopes: No

Auth appears to be passed as a single configured header value; there is no indication of OAuth2 or fine-grained scope enforcement.

Pricing

Free tier: No
Requires CC: No

No pricing information provided; this appears to be an open-source/self-hosted MCP server.

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • Tool names depend on operationId; missing/duplicate operationId may lead to less stable/less predictable tool naming.
  • The server accepts OpenAPI JSON only (YAML not supported), so specs may need conversion.
  • Static header auth (-auth) may not work for APIs requiring OAuth2 token exchange/refresh flows.
  • Rate limiting behavior is not documented; agents may need to implement backoff outside the MCP layer.
  • Spec completeness matters: incorrect parameter schemas or missing required fields can cause tool call failures.

Alternatives

Full Evaluation Report

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

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 2026-04-04.

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