Navigate code structure via AST, trace impact and data flow, chain complex tasks, and recall project context. Built for developers who need deep code analysis and multi-step refactoring workflows.
io.github.halflifezyf2680/mpm-coding
Local install
STDIO
No auth required
How models use it and what it is built for.
Navigate code structure via AST, trace impact and data flow, chain complex tasks, and recall project context. Built for developers who need deep code analysis and multi-step refactoring workflows.
Local install — runs as a subprocess.
Where to find authoritative docs and source for MPM-Coding.
Paste any of these into Agent Studio after connecting MPM-Coding.
Common questions about connecting and running MPM-Coding.
What does AST navigation mean in this MCP server?
The server parses your code into an Abstract Syntax Tree and lets you explore structure, find definitions, and trace relationships between functions and variables. This is faster than grep for understanding code architecture.
How does impact/flow analysis work?
It traces how changes to one function or variable ripple through your codebase—showing you which other functions depend on it and what will break if you modify it.
What are task chains?
Task chains let you define multi-step refactoring or analysis workflows (e.g., rename function → update imports → run linter) and execute them as a single operation instead of manually running each step.
How does the memo/recall feature work?
The server stores context about your project—schemas, dependencies, past decisions—so you can reference them later without re-explaining your codebase architecture in each conversation.
What language support does this server have?
Check the GitHub releases page and documentation for the full list of supported languages. The AST-based approach typically covers JavaScript, Python, Java, and other mainstream languages.
MCP Playground runs 10,000+ hosted MCP servers — GitHub, Linear, Notion, Stripe, Sentry and more — across Claude, GPT, Gemini, DeepSeek and 30+ AI models. Compare model answers side-by-side, save agent presets, share runs. Zero install.
Open Agent Studio