MCP Chaining Pipeline A research to implement pipeline that chains 5 MCP tools for end to end workflows. When to Use Building multi tool MCP pipelines Understanding how to chain MCP calls with graceful degradation Debugging MCP environment variable issues Learning the tool naming conventions for different MCP servers What We Built A pipeline that chains these tools: Step Server Tool ID Purpose 1 nia nia search Search library documentation 2 ast grep ast grep find code Find AST code patterns 3 morph morph warpgrep codebase search Fast codebase search 4 qlty qlty qlty check Code quality validation 5 git git git status Git operations Key Files scripts/research implement pipeline.py Main pipeline implementation scripts/test research pipeline.py Test harness with isolated sandbox workspace/pipeline test/sample code.py Test sample code Usage Examples Critical Fix: Environment Variables The MCP SDK's get default environment() only includes basic vars (PATH, HOME, etc.), NOT os.environ . We fixed src/runtime/mcp client.py to pass full environment: This ensures API keys from ~/.claude/.env reach subprocesses. Graceful Degradation Pattern Each tool is optional. If unavailable (disabled, no API key, etc.), the pipeline continues: Tool Name Reference nia (Documentation Search) ast grep (Structural Code Search) morph (Fast Text Search + Edit) qlty (Code Quality) git (Version Control) Pipeline Architecture Error Handling The pipeline captures errors without failing the entire run: Creating Your Own Pipeline 1. Copy the pattern from scripts/research implement pipeline.py 2. Define your steps as async functions 3. Use check tool available() for graceful degradation 4. Chain results through PipelineContext 5. Aggregate with print summary()

mcp-chaining