@suncreation/opencode-toolsearch
MCP ServerFreeMulti-provider request patch, Anthropic OAuth bridge, and MCP tool discovery for OpenCode
Capabilities5 decomposed
multi-provider http request patching for llm api compatibility
Medium confidenceIntercepts and patches HTTP requests at the transport layer to normalize API calls across multiple LLM providers (OpenAI, Anthropic, GLM, etc.). Uses a provider-agnostic request/response transformation pipeline that maps provider-specific schemas to a unified interface, enabling seamless provider switching without changing application code. Patches are applied at the Node.js HTTP module level, intercepting requests before they reach provider endpoints.
Implements transport-layer request patching rather than SDK wrapping, allowing provider switching without dependency on provider-specific SDKs or client libraries. Patches Node.js HTTP module directly to intercept and transform requests before they leave the application.
More transparent than wrapper SDKs because it operates at the HTTP layer, enabling existing code using native fetch/axios to work with multiple providers without refactoring.
anthropic oauth bridge with token exchange and session management
Medium confidenceImplements OAuth 2.0 authorization flow for Anthropic API access, handling token exchange, refresh token rotation, and session lifecycle management. Bridges between OAuth identity providers and Anthropic's authentication system, storing and rotating credentials securely. Manages token expiration, automatic refresh, and fallback to API key authentication when OAuth tokens are unavailable.
Provides native OAuth bridge specifically for Anthropic rather than generic OAuth handling, with built-in understanding of Anthropic's token formats, expiration windows, and refresh semantics. Includes automatic fallback to API key authentication for hybrid scenarios.
Purpose-built for Anthropic OAuth unlike generic OAuth libraries, reducing boilerplate and handling Anthropic-specific token lifecycle quirks automatically.
mcp server discovery and tool registry with provider-aware routing
Medium confidenceDiscovers and catalogs available Model Context Protocol (MCP) servers and their exposed tools, building a dynamic registry that maps tool names to server endpoints and capabilities. Uses MCP protocol introspection to query server metadata, tool schemas, and supported operations. Routes tool invocations to the correct MCP server based on tool name, provider affinity, or capability matching. Maintains a cached registry to avoid repeated discovery overhead.
Implements dynamic MCP tool discovery with provider-aware routing rather than static tool configuration, using MCP protocol introspection to build registries at runtime. Includes caching and fallback mechanisms for resilience across multiple MCP servers.
Eliminates manual tool registration by auto-discovering MCP servers and their capabilities, whereas most MCP integrations require explicit tool lists in configuration files.
opencode integration bridge for tool discovery and execution
Medium confidenceBridges OpenCode development environment with MCP tool discovery and multi-provider LLM support, exposing discovered tools as OpenCode extensions. Translates between OpenCode's tool invocation model and MCP server protocols, handling argument marshaling, error handling, and result formatting. Enables OpenCode to dynamically load tools from MCP servers without hardcoded tool lists.
Provides first-class OpenCode IDE integration for MCP tools, translating between OpenCode's extension model and MCP protocols. Enables dynamic tool loading in OpenCode without requiring IDE restart or manual extension installation.
OpenCode-native integration versus generic MCP clients, providing seamless IDE experience with native UI rendering and workflow integration.
glm provider support with request/response transformation
Medium confidenceExtends multi-provider request patching to support Zhipu AI's GLM API, implementing request schema translation from OpenAI/Anthropic formats to GLM's proprietary API contract. Handles GLM-specific features (model variants, parameter mappings, response formats) and error codes. Transforms GLM responses back to normalized format for downstream consumption by application code.
Implements GLM-specific request/response transformation as part of multi-provider abstraction, handling GLM's unique parameter mappings and response formats. Includes fallback handling for GLM-unsupported features.
Enables GLM usage in provider-agnostic code without separate GLM SDK dependency, whereas most applications require GLM-specific integration code.
Capabilities are decomposed by AI analysis. Each maps to specific user intents and improves with match feedback.
Related Artifactssharing capabilities
Artifacts that share capabilities with @suncreation/opencode-toolsearch, ranked by overlap. Discovered automatically through the match graph.
AgentR Universal MCP SDK
** - A python SDK to build MCP Servers with inbuilt credential management by **[Agentr](https://agentr.dev/home)**
@auto-engineer/ai-gateway
Unified AI provider abstraction layer with multi-provider support and MCP tool integration.
Portkey
A full-stack LLMOps platform for LLM monitoring, caching, and management.
OpenRouter
A unified interface for LLMs. [#opensource](https://github.com/OpenRouterTeam)
@clerk/mcp-tools
Tools for writing MCP clients and servers without pain
gpt-computer-assistant
** dockerized mcp client with Anthropic, OpenAI and Langchain.
Best For
- ✓Teams building multi-provider LLM applications
- ✓Developers prototyping provider-agnostic AI agents
- ✓Organizations evaluating multiple LLM vendors
- ✓SaaS platforms offering Claude API access to multiple users
- ✓Enterprise applications requiring OAuth-based authentication
- ✓Teams implementing zero-trust credential management
- ✓Developers building MCP-based agent systems with multiple tool providers
- ✓Teams deploying heterogeneous tool ecosystems (local tools, cloud APIs, custom servers)
Known Limitations
- ⚠Patches applied at runtime — cannot optimize request serialization at compile time
- ⚠Provider-specific features (streaming, vision, function calling variants) may require conditional logic despite normalization
- ⚠Patching at HTTP module level adds ~5-15ms overhead per request due to transformation pipeline
- ⚠Requires Anthropic OAuth provider registration — not available for all Anthropic accounts
- ⚠Token refresh adds ~200-500ms latency on first request after expiration
- ⚠No built-in credential encryption — requires external secret management (e.g., AWS Secrets Manager, HashiCorp Vault)
Requirements
Input / Output
UnfragileRank
UnfragileRank is computed from adoption signals, documentation quality, ecosystem connectivity, match graph feedback, and freshness. No artifact can pay for a higher rank.
Package Details
About
Multi-provider request patch, Anthropic OAuth bridge, and MCP tool discovery for OpenCode
Categories
Alternatives to @suncreation/opencode-toolsearch
Are you the builder of @suncreation/opencode-toolsearch?
Claim this artifact to get a verified badge, access match analytics, see which intents users search for, and manage your listing.
Get the weekly brief
New tools, rising stars, and what's actually worth your time. No spam.
Data Sources
Looking for something else?
Search →