{"id":"mcp360-unified-gateway-mcp","name":"unified-gateway-mcp","homepage":"https://mcp360.ai","repo_url":"https://github.com/mcp360/unified-gateway-mcp","category":"devtools","subcategories":[],"tags":["mcp","agent-tools","gateway","ai-agents","tool-discovery","tool-execution","typescript","marketplace","oauth","api-key"],"what_it_does":"Provides an MCP (Model Context Protocol) gateway that aggregates many external MCP tools/services behind two meta-tools—`search` to discover tools and `execute` to run a selected tool—with authentication via API key or OAuth2.","use_cases":["Search and select from a large marketplace of MCP tools","Dynamically run heterogeneous actions (e.g., SEO, email verification, web scraping) through a single gateway","Build agent workflows (Claude/Cursor/VS Code integrations) without integrating each underlying API separately","Use/attach custom MCP servers via the same unified gateway interface"],"not_for":["High-assurance environments requiring formally specified SLAs/error contracts that are not documented in the README","Workloads where you need direct, first-party access to specific APIs instead of an aggregated abstraction","Use cases requiring strict, documented data residency or compliance guarantees beyond marketing claims (not evidenced in provided text)"],"best_when":"You want agents to access many MCP-backed tools with minimal integration effort and can authenticate to MCP360.","avoid_when":"You require comprehensive, versioned API contracts (OpenAPI/SDK), explicit rate-limit headers, or documented idempotency/retry behavior for reliability-critical operations.","alternatives":["Directly integrate the underlying MCP servers (or their native APIs) instead of using a gateway","Use an MCP client/registry solution that exposes well-documented REST/OpenAPI endpoints","Build a small custom MCP gateway tailored to the specific set of tools you need"],"af_score":54.5,"security_score":66.8,"reliability_score":22.5,"package_type":"mcp_server","discovery_source":["github"],"priority":"high","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-03-30T15:34:42.498502+00:00","interface":{"has_rest_api":false,"has_graphql":false,"has_grpc":false,"has_mcp_server":true,"mcp_server_url":"https://connect.mcp360.ai/v1/mcp360/mcp","has_sdk":false,"sdk_languages":[],"openapi_spec_url":null,"webhooks":false},"auth":{"methods":["API key (MCP360_API_KEY)","OAuth 2.0 (browser sign-in fallback when API key not provided)"],"oauth":true,"scopes":false,"notes":"README describes API key configuration and an OAuth2 automatic fallback flow. Provided text does not document exact scopes/permissions granularity for OAuth."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":[],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"Pricing details and limits are not present in the provided README excerpt."},"requirements":{"requires_signup":true,"requires_credit_card":false,"domain_verification":false,"data_residency":[],"compliance":["GDPR (claimed)","SOC 2 Type II (claimed)","ISO 27001 (claimed)"],"min_contract":null},"agent_readiness":{"af_score":54.5,"security_score":66.8,"reliability_score":22.5,"mcp_server_quality":55.0,"documentation_accuracy":65.0,"error_message_quality":0.0,"error_message_notes":null,"auth_complexity":70.0,"rate_limit_clarity":45.0,"tls_enforcement":95.0,"auth_strength":80.0,"scope_granularity":35.0,"dependency_hygiene":45.0,"secret_handling":70.0,"security_notes":"The README claims TLS 1.3, OAuth token refresh, audit logs, rate limiting, DDoS protection, and anomaly detection, but the provided text does not show concrete implementation details. API keys are passed via environment variables in client configs (better than embedding in code), but explicit guidance on secret storage/redaction and token scope minimization is not evidenced.","uptime_documented":20.0,"version_stability":25.0,"breaking_changes_history":20.0,"error_recovery":25.0,"idempotency_support":"false","idempotency_notes":null,"pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["`execute` is a dynamic router; if tool parameters don’t match the selected tool’s schema, errors may be opaque because schema validation/error formats are not shown in the provided text.","Search results may vary based on connected marketplace/custom MCP availability, so an agent may need to re-search before executing in long-running sessions.","No documented retry/idempotency semantics for `execute`, so agents should avoid blind retries for side-effecting tools."]}}