Big Indexer
Give Claude or Cursor architecture-aware repo context—module boundaries, seams, and blast-radius—before you ask the agent to refactor or add features.
Overview
Big Indexer is a MCP server for the Build phase that supplies architecture-aware context—boundaries, seams, impact scope, and BGI-TWIN guidance—to coding agents.
What is this MCP server?
- Architecture-aware MCP context: boundaries, seams, and impact scope for safer edits
- BGI-TWIN guidance for reasoning about change across the codebase
- PyPI package `bigindexer` v0.1.2 with stdio transport
- GitHub source at ahmedxuhri/bigindexer for self-host or inspect
- Designed to reduce cross-boundary mistakes during AI-assisted refactors
- Server version 0.1.2 on PyPI identifier `bigindexer`
- Transport: stdio
- Repository: github.com/ahmedxuhri/bigindexer
Community signal: 2 GitHub stars.
What problem does it solve?
Agents that only see flat file snippets miss module boundaries and blast radius, so one “small” prompt can break seams you did not know existed.
Who is it for?
Indie developers with an existing multi-module or service-shaped codebase who use Claude Code or Cursor daily and want richer structural context in MCP.
Skip if: Empty repos, pure content sites with no architecture story, or teams that forbid local MCP servers on laptops.
What do I get? / Deliverables
After you register Big Indexer over stdio, the agent can reason with explicit boundaries and impact scope so refactors and feature work stay inside the right architectural seams.
- Registered stdio MCP server exposing architecture-aware context tools
- Agent sessions that reference boundaries, seams, and impact scope
- BGI-TWIN-oriented guidance available during implementation prompts
Recommended MCP Servers
Journey fit
Solo builders lean on agents during implementation; structured architecture context belongs in the build phase when code and structure change fastest. Big Indexer is wired as MCP stdio tooling that enriches agent sessions, not a standalone app feature—canonical shelf is agent-tooling.
How it compares
MCP architecture-context server, not a linter skill or a hosted code-search SaaS by itself.
Common Questions / FAQ
Who is Big Indexer for?
Solo builders and small teams using AI coding agents on real codebases who need boundaries, seams, and impact scope in context—not just file paths.
When should I use Big Indexer?
Use it during build work when you are refactoring, adding integrations, or planning changes that could cross module or service boundaries.
How do I add Big Indexer to my agent?
Install the PyPI package `bigindexer` (v0.1.2), add an MCP server entry with stdio transport pointing at that package, restart Claude Code or your MCP-capable client, and verify tools load from the registry manifest.