agency-lsp-index-engineer

Orchestrate multiple LSP clients into a unified semantic code graph.

Updated Feb 11, 2026
One-click install
npx skills add https://github.com/augustoheiss/LogicDefense --skill agency-lsp-index-engineer-augustoheiss
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agency-lsp-index-engineer
Source: https://github.com/augustoheiss/LogicDefense/tree/main/.gemini/skills/agency-lsp-index-engineer
Command: npx skills add https://github.com/augustoheiss/LogicDefense --skill agency-lsp-index-engineer-augustoheiss

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Projects with multiple languages and independent language servers suffer from fragmented code intelligence, slow cross-references, and inconsistent navigation results; this Skill unifies heterogeneous LSP clients into a single semantic graph so editors and web UIs can deliver fast, consistent go-to-definition, hover, and reference queries across a polyglot codebase.

Core Features & Use Cases

  • Multi-LSP Orchestration: Initialize and manage concurrent LSP clients (TypeScript, PHP, Go, Rust, Python) with proper capability negotiation and lifecycle management.
  • Semantic Graph Construction: Transform LSP responses into a unified graph of file and symbol nodes with typed edges for contains/imports/calls/references and maintain strict graph consistency.
  • Real-time Incremental Indexing: Produce nav.index.jsonl and LSIF import/export artifacts, stream graph diffs via WebSocket, and persist caches in SQLite/JSON for fast startup and incremental updates.
  • Performance & Scale Targets: Designed to meet sub-100ms lookup goals, aggressive but precise caching, batching of LSP requests, and strategies for handling 25k–100k+ symbols with low memory overhead.
  • Use Case: Power a web-based code explorer that provides instant cross-language navigation and live symbol updates after commits or file saves.

Quick Start

Start the LSP orchestrator on your project root to initialize TypeScript and PHP language servers, build the semantic graph, and generate a nav.index.jsonl navigation index.

Frequently Asked Questions about agency-lsp-index-engineer

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

FAQPage Schema
How do I unify multiple language servers into a single semantic code graph for cross-language navigation?

To unify multiple language servers into a semantic code graph, orchestrate concurrent LSP clients like TypeScript and PHP to transform responses into a graph of file and symbol nodes with typed edges for deterministic cross-language go-to-definition and reference queries.

What is the best way to achieve real-time incremental indexing for a polyglot codebase?

Real-time incremental indexing for a polyglot codebase is achieved by streaming graph diffs via WebSocket, batching LSP requests, and persisting caches in SQLite or JSON to enable fast startup and live symbol updates after file saves.

How do I export a navigation index to LSIF format from heterogeneous LSP clients?

Export a navigation index to LSIF format from heterogeneous LSP clients by transforming their responses into a unified semantic graph and generating nav.index.jsonl files alongside LSIF import/export artifacts for consistent code intelligence.

Can I maintain sub-100ms lookup performance for a code graph with over 100k symbols?

You can maintain sub-100ms lookup performance for a code graph with 100k symbols by applying aggressive but precise caching, batching LSP requests, and utilizing SQLite or JSON persistence strategies designed for low memory overhead.

Does this multi-LSP orchestration approach work with TypeScript, PHP, Go, Rust, and Python environments?

This multi-LSP orchestration approach works with TypeScript, PHP, Go, Rust, and Python environments by initializing concurrent LSP clients with proper capability negotiation and lifecycle management to ensure consistent navigation across the polyglot repository.