{"id":"joecastrom-mcp-chat-studio","name":"mcp-chat-studio","homepage":null,"repo_url":"https://github.com/JoeCastrom/mcp-chat-studio","category":"devtools","subcategories":[],"tags":["mcp","model-context-protocol","testing","debugging","ui","llm","openapi","contract-testing","workflows"],"what_it_does":"Local, UI-based MCP (Model Context Protocol) testing and development platform. Provides classic/workspace layouts, lets you connect MCP servers (STDIO/SSE), inspect JSON-RPC traffic, run test scenarios (record/replay, bulk testing, diffing), and generate documentation and workflow exports (Python/Node). Includes multi-provider LLM chat support (e.g., Ollama/OpenAI/Claude/Gemini and others) to assist with workflow building and generation.","use_cases":["Test and debug MCP servers end-to-end (tool calling, schema validation, response diffs)","Create and replay recorded tool-call scenarios; bulk and matrix testing across multiple MCP servers","Generate documentation bundles and exported workflow scripts (Python/Node) from visual flows","Regression/flake detection via health metrics, baselines, and contract testing concepts","Prototyping MCP tool schemas by importing OpenAPI and generating MCP proxy servers/tools"],"not_for":["Production, internet-exposed services (explicitly positioned as a local dev test bench)","Acts as a managed hosted API for others to call (it’s primarily an app/desktop/web UI running locally)","Highly regulated environments without reviewing its auth/session storage and dependency/security posture"],"best_when":"You need a local development/testing cockpit for MCP servers with visual debugging, scenario replay, and cross-server comparisons, optionally assisted by LLMs.","avoid_when":"You need a strictly documented external API for programmatic integration by other services/agents, or you require strong, externally enforced access controls as a hosted multi-tenant service.","alternatives":["Claude Desktop/other MCP clients with limited testing features","Custom MCP test harnesses using the MCP SDK directly","OpenAPI-to-MCP generator tooling (if you only need generation, not testing/diffing UI)","General-purpose JSON-RPC tracing tools plus bespoke scenario runners"],"af_score":32.8,"security_score":41.8,"reliability_score":38.8,"package_type":"mcp_server","discovery_source":["github"],"priority":"high","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-03-30T15:35:09.820302+00:00","interface":{"has_rest_api":false,"has_graphql":false,"has_grpc":false,"has_mcp_server":true,"mcp_server_url":null,"has_sdk":true,"sdk_languages":["Python","Node.js","TypeScript"],"openapi_spec_url":null,"webhooks":false},"auth":{"methods":["OAuth (UI configuration mentioned)","Bearer/Basic/API-key and header/query overrides (per-request auth overrides mentioned)","Environment-variable based API keys per server"],"oauth":true,"scopes":false,"notes":"Auth is described as UI-configurable for LLM/provider settings and per-request overrides for OpenAPI proxy tools. No explicit scope model is documented in the provided content."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":[],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"Open-source (MIT) and used locally; any LLM costs depend on configured provider(s), but pricing for a hosted service is not indicated."},"requirements":{"requires_signup":false,"requires_credit_card":false,"domain_verification":false,"data_residency":[],"compliance":[],"min_contract":null},"agent_readiness":{"af_score":32.8,"security_score":41.8,"reliability_score":38.8,"mcp_server_quality":55.0,"documentation_accuracy":55.0,"error_message_quality":0.0,"error_message_notes":null,"auth_complexity":55.0,"rate_limit_clarity":25.0,"tls_enforcement":30.0,"auth_strength":55.0,"scope_granularity":20.0,"dependency_hygiene":60.0,"secret_handling":45.0,"security_notes":"Positioned as a local dev tool (not for internet exposure). Dependencies include common web/security libraries (e.g., express-rate-limit, vm2, cors), but the provided content does not confirm TLS enforcement for all interfaces, nor does it document secret-handling/logging guarantees or fine-grained scope controls. Per-request auth overrides and OAuth UI suggest auth capabilities, but scope granularity and security boundaries are not clearly specified in the available text.","uptime_documented":0.0,"version_stability":60.0,"breaking_changes_history":50.0,"error_recovery":45.0,"idempotency_support":"false","idempotency_notes":null,"pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["Primary usage is via a browser UI rather than a clearly documented external agent-facing API; automated agents may need UI driving or reverse-engineering of internal endpoints.","Local-only positioning means availability/transport assumptions (STDIO/SSE) must be handled by the agent as part of setup.","Idempotency semantics for replay/bulk operations are not documented in the provided materials; agents should avoid assuming safe retries for state-changing tool calls."]}}