Microsoft Research (with UC Berkeley and Microsoft Security) · Coding Agent Plugin

TokenEconomizer

TokenEconomizer is an open-source plugin that learns from a coding agent's past sessions in a repository to cut token spend, without modifying the agent or its underlying model. It uses three mechanisms: repository-specific micro-skills, pruning of unused tools, and failure-loop intervention. Across 277 SWE-Bench tasks from six repositories, mean cost fell 18.4–30.9% across five agent and model combinations, with task success rates broadly stable. The plugin needs no API key of its own and keeps what it learns on the developer's machine.

Open the source ↗

Similar Agents & Automation releases

  • ESCAPEMENT

    Leonxlnx · Agent-made Blender film · Week 47, 2026

  • Cindy

    makecindy · Multi-device agent client · Week 47, 2026

  • OpenMontage

    OpenMontage (Calesthio AI Labs) · Agentic video studio · Week 47, 2026

  • Antigravity /teamwork

    Google Research · Multi-agent Research Harness · Week 47, 2026

  • numbrone

    Steven Michael Thomas (community project built on herdr) · Agent Orchestration · Week 47, 2026

  • Take One

    talsiach (Tal Siach) · Video creation skill · Week 46, 2026

Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.