Generate multi-layer LLM-friendly indexes for Rust codebases.
npx skills add https://github.com/zfedoran/rsmap --skill rsmap
Please help me install this Agent Skill. Skill: rsmap Source: https://github.com/zfedoran/rsmap/tree/main Command: npx skills add https://github.com/zfedoran/rsmap --skill rsmap
This Skill helps you understand a Rust project without expensive full-source scanning by using precomputed, multi-layer indexes designed for LLM discovery and navigation.
overview.md, api-surface.md, relationships.md, and index.json to support progressively deeper understanding.index.json to map fully-qualified item paths to file locations and line ranges for targeted reading.cache.json and hashes to avoid unnecessary re-parsing while still updating annotations staleness.Ask the AI to run rsmap generate with your Rust project path and an output directory to produce the layered index files.
High-intent search queries and answers about installing and using this skill.
Mapping Rust codebase architecture for LLM semantic search involves generating precomputed, multi-layer indexes. This produces overview, API surface, and relationship files that enable efficient discovery without full-source scanning.
The best way to discover API surface and module relationships in a Rust project is by generating layered index files like api-surface.md and relationships.md. These provide progressively deeper understanding of crates, modules, and item signatures.
You locate exact source file locations and line ranges for Rust items using the generated index.json file. It maps fully-qualified item paths to specific file locations and line ranges for targeted reading by an LLM.
You can add custom LLM descriptions to indexed Rust items and track staleness by exporting unannotated items to annotations.toml. Fill in notes externally, import them back, and use cache.json with hashes to track staleness during incremental rebuilds.
Generating Rust codebase indexes for LLMs does not require full rebuilds every time. The process supports incremental rebuilds using cache.json and hashes to avoid unnecessary re-parsing while still updating annotation staleness.
open-haxIndex OpenCode sessions into a Chroma store and generate reconstruction notes.
scooter-lacroixExplore codebases and search code with configurable context via CLI.
simorgh3196Evaluate Rust crates across serialization, async runtimes, HTTP frameworks, and databases.
kayzaaAutomate tax loss harvesting with wash sale prevention for financial portfolios.