ecomode

Optimizes agent-workflow task routing across LOW, MEDIUM, and HIGH tiers to conserve tokens.

Updated Apr 10, 2026
One-click install
npx skills add https://github.com/lpjhelder/kronon-forge --skill ecomode-lpjhelder
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ecomode
Source: https://github.com/lpjhelder/kronon-forge/tree/main/skills/ecomode
Command: npx skills add https://github.com/lpjhelder/kronon-forge --skill ecomode-lpjhelder

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Ecomode minimizes token consumption and inference costs by steering work toward lower-cost model tiers during agent orchestration, reducing unnecessary use of expensive thorough models while preserving quality for critical tasks.

Core Features & Use Cases

  • Cost-aware model routing: Prefers LOW-tier agents for most tasks, falls back to STANDARD only when LOW fails, and reserves THOROUGH for essential planning or critique.
  • Agent delegation guidance: Provides explicit delegation preferences for executor, explore, architect, planner, and writer roles to optimize cost without changing orchestration rules.
  • Operational controls: Supports background execution limits, lifecycle state via forge_state, and can be disabled in user config for full control.
  • Use Case: Run large-scale code discovery, documentation, or routine refactors across a repository while significantly reducing token usage and cost.

Quick Start

Activate ecomode in your Forge session to route tasks to LOW-tier agents and reduce token consumption.

Frequently Asked Questions about ecomode

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

FAQPage Schema
How do I reduce token consumption during agent orchestration?

Reduce token consumption during agent orchestration by applying smart model routing that prefers LOW-tier agents for routine tasks and escalates to STANDARD or THOROUGH tiers only when low-cost models fail or require multi-file reasoning.

How does cost-aware model routing work for code analysis and refactoring?

Cost-aware model routing assigns code analysis and refactoring tasks to LOW-tier agents first, escalating to STANDARD only upon failure or when complex multi-file reasoning is needed, reserving THOROUGH models for essential planning or critique.

Can I use tier-based agent delegation with Forge and Kronon workflows?

Yes, tier-based agent delegation integrates with Forge and Kronon workflows by enforcing LOW-tier preferences across executor, explore, architect, planner, and writer roles while relying on forge_state for lifecycle decision tracking.

Do I need agent-tiers references to optimize model selection for large-scale code discovery?

Agent-tiers references provide the selection guidance needed to optimize model selection for large-scale code discovery, ensuring tasks route to LOW-tier agents first to minimize inference costs without changing orchestration rules.

What is the best way to control token costs when running repository-wide documentation tasks?

Control token costs for repository-wide documentation tasks by activating cost-optimization routing that delegates to LOW-tier agents by default, escalating only when task complexity demands STANDARD or THOROUGH model tiers.

Can I disable smart model routing if a task requires expensive thorough models?

Yes, smart model routing can be disabled through user configuration to allow full manual control over agent selection, bypassing the automatic LOW-tier preference and lifecycle state enforcement when critical tasks demand higher-tier models.