Context Engineering Prompt Optimizer

1. System Prompt & Scenario Config Tokens: ~0

Target Model

Token Budget: 8,000

Monolithic System Prompt Input (13 Rules Default)

Simulated User Query / Context

2. Optimization Strategies

Prune Generic & Passive Negative Rules Removes generic boilerplate (PEP8, markdown format, verbose warnings) inherent to modern LLMs.

Delegate Architecture & Repo Rules to CLAUDE.md Moves static linters, git workflow, and style conventions into repository context.

Modularize Workflows into Dynamic Skills Extracts specialized execution steps (Summaries, Security Checks) into triggers.

3. Live Context Efficiency Analytics Optimized

System Tokens

-0%

Attention Overhead

Low

Instruction Density

0.0

Rules/100w

Budget Efficiency

Saved

Refactoring Log & Action Recipes