agency-lsp-index-engineer

Orchestrate multiple LSP clients into a unified semantic graph for polyglot repositories.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill solves the fragmentation of code intelligence across multiple language servers by orchestrating LSP clients and consolidating their outputs into a single, consistent semantic graph that powers fast go-to-definition, references, and hover navigation.

Core Features & Use Cases

  • Multi-Language Orchestration: Initialize and manage TypeScript, PHP, Go, Rust, and Python language servers in parallel with proper capability negotiation and lifecycle management.
  • Semantic Indexing & Navigation: Produce nav.index.jsonl and LSIF-compatible exports that record symbol definitions, references, and hover content for fast lookup and offline analysis.
  • Real-time Incremental Updates: Maintain an in-memory graph with atomic diffs streamed over WebSocket, support file watchers and git hooks, and ensure strict one-definition-per-symbol consistency for reliable navigation.
  • Performance Targets: Designed to meet low-latency contracts (sub-500ms updates, sub-100ms graph queries for small datasets) with batching, caching, and scalable memory strategies.

Quick Start

Initialize TypeScript and PHP language servers for your project root and generate a nav.index.jsonl while streaming graph diffs over WebSocket.

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?

To unify multiple language servers into a single semantic code graph, this Skill orchestrates TypeScript, PHP, Go, Rust, and Python LSP clients in parallel. It consolidates symbols, references, and hover data into one consistent graph for reliable navigation.

How do I generate an LSIF index and nav.index.jsonl from a polyglot repository?

You generate an LSIF index and nav.index.jsonl from a polyglot repository by extracting symbol definitions, references, and hover content through orchestrated LSP clients. The Skill supports LSIF import/export for offline analysis and fast lookup.

Can I get real-time incremental code intelligence updates over WebSocket?

Yes, you can get real-time incremental code intelligence updates over WebSocket. The Skill maintains an in-memory graph with atomic diffs, supports file watchers and git hooks, and targets sub-500ms updates for live code navigation.

Does this approach ensure single-definition node guarantees for LSP 3.17 compliance?

Yes, this approach ensures single-definition node guarantees for LSP 3.17 compliance. It enforces strict one-definition-per-symbol consistency during semantic indexing to provide reliable go-to-definition and references navigation.

What is the best way to query a semantic code graph with low latency?

The best way to query a semantic code graph with low latency is through the provided HTTP and WS endpoints. The Skill uses batching, caching, and scalable memory strategies to achieve sub-100ms graph queries for small datasets.