lsp-simulate

Simulate LSP-based code edits in memory before applying them to disk.

102|5|Updated Apr 6, 2026
One-click install
npx skills add https://github.com/blackwell-systems/agent-lsp --skill lsp-simulate
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: lsp-simulate
Source: https://github.com/blackwell-systems/agent-lsp/tree/main/skills/lsp-simulate
Command: npx skills add https://github.com/blackwell-systems/agent-lsp --skill lsp-simulate

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Simulate code edits in memory before applying changes to disk, enabling safe planning and validation of multi-file edits without touching live files.

Core Features & Use Cases

  • In-memory simulation of edits with real-time diagnostics from the LSP overlay.
  • Stepwise evaluation and optional commit or discard to produce a safe patch.
  • Support for chained edits and workspace-wide planning across multiple files.

Quick Start

Create a simulation session, apply edits in memory, then evaluate the session to decide whether to commit.

Frequently Asked Questions about lsp-simulate

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

FAQPage Schema
How do I safely validate multi-file code edits before writing to disk?

In-memory LSP simulation applies code edits to an overlay rather than disk, allowing you to validate multi-file refactors with real-time diagnostics. You can evaluate the session stepwise and choose to commit or discard the resulting patch safely.

Can I test risky workspace refactors without modifying live files?

Yes, you can test risky workspace refactors without modifying live files by running an in-memory LSP simulation. The workflow applies edits to an overlay, evaluates diagnostics stepwise, and discards the session if errors occur, keeping your workspace untouched.

Do I need an agent-lsp server to run in-memory edit simulations?

Yes, an agent-lsp server is required to run in-memory edit simulations. It provides the diagnostic overlay necessary to evaluate code changes in memory and supports the start, simulate, evaluate, commit, and destroy workflow operations.

What is the best way to plan chained code edits across multiple files?

The best way to plan chained edits across multiple files is using an in-memory simulation session. This workflow supports applying sequential edits, evaluating real-time LSP diagnostics at each step, and committing the validated patch only when ready.

How does stepwise evaluation work for simulated code refactoring?

Stepwise evaluation in simulated code refactoring works by applying edits to an in-memory LSP overlay and checking diagnostics after each step. You can review errors incrementally, deciding whether to continue chaining edits, commit the patch, or destroy the session.