rmc-codemap

Generate task-conditioned Rust workspace subgraphs from a persisted HIR hypergraph.

29|5|Updated Nov 24, 2025
One-click install
npx skills add https://github.com/molaco/rust-code-mcp --skill rmc-codemap
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rmc-codemap
Source: https://github.com/molaco/rust-code-mcp/tree/main/skills/rmc-codemap
Command: npx skills add https://github.com/molaco/rust-code-mcp --skill rmc-codemap

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Codemap helps you quickly answer “what parts of this Rust workspace matter for my task?” by extracting a relevant, task-shaped subgraph from a much larger code hypergraph.

Core Features & Use Cases

  • Task-conditioned subgraphs: Build a BFS-expanded, relevance-scored workspace slice from either a natural-language prompt or explicit seed qualified names.
  • Hierarchy + graph export formats: Get results as json, outline, mermaid, or all views for different consumption needs.
  • Optional snippet inclusion: Attach small source excerpts to the nodes for faster inspection and LLM grounding.
  • Staleness diagnostics and guardrails: Automatically surface when the hypergraph snapshot is older than the newest Rust sources, so you can rebuild before trusting results.

Quick Start

Ask for “build_codemap for what touches module X” and set the seed to the qualified name of the public item you want to analyze.

Frequently Asked Questions about rmc-codemap

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract a focused code subgraph from a large Rust workspace?

To extract a Rust workspace subgraph, provide a natural-language prompt or explicit seed qualified names. The tool performs BFS expansion over a persisted HIR hypergraph, applying relevance scoring to generate a bounded, task-conditioned slice of your codebase.

Can I generate mermaid diagrams for Rust symbol navigation?

Yes, you can generate mermaid diagrams for Rust symbol navigation by selecting the mermaid export format. The tool projects hierarchy and graph relationships into mermaid syntax, alongside options for json, outline, or all views for different documentation needs.

Does semantic code exploration work with natural language prompts?

Semantic code exploration supports natural language prompts through the task_prompt mode, which requires hybrid search readiness via index_codebase. This allows you to build a relevance-scored workspace slice from a task description rather than explicit seed names.

What's the best way to map subsystem dependencies in a Rust codebase?

The best way to map subsystem dependencies is using cohesion and dependency analysis features. By extracting a task-shaped subgraph from the workspace hypergraph, you get a bounded view of relevant nodes and edges rather than analyzing a single overwhelming call graph.

Why does my Rust code exploration show staleness diagnostics?

Staleness diagnostics appear when the persisted hypergraph snapshot is older than the newest Rust source files. This guardrail surfaces automatically, warning you to rebuild the hypergraph before trusting the extracted subgraph results for navigation or analysis.

Do I need to build a hypergraph before generating a code subgraph?

Yes, you must run build_hypergraph to create a directory snapshot before generating a code subgraph. This persisted HIR hypergraph is required for BFS expansion, and index_codebase is additionally needed for task_prompt mode hybrid search readiness.