Coordinate decentralized job allocation among autonomous agents using reputation-weighted Thompson sampling.
npx skills add https://github.com/kriscendobot/garden --skill bid-auction
Please help me install this Agent Skill. Skill: bid-auction Source: https://github.com/kriscendobot/garden/tree/main/skills/bid-auction Command: npx skills add https://github.com/kriscendobot/garden --skill bid-auction
This skill solves the problem of inefficient and centralized job distribution in a fleet of autonomous agents by enabling a deterministic, reputation-based auction system that selects the most cost-effective worker for any given task.
Enable the bid-auction skill by setting the environment variable GARDEN_QUOTA_ROUTING to auction in your worker configuration.
High-intent search queries and answers about installing and using this skill.
Decentralized job allocation among autonomous agents is coordinated using reputation-weighted Thompson sampling. This approach evaluates cost-per-acceptance metrics and historical performance data to distribute tasks efficiently across distributed worker fleets without a central authority.
Reputation-weighted Thompson sampling is a mechanism that resolves job awards by tracking worker performance and cost metrics. It favors efficient, high-quality implementations over simple token-cost metrics to select the most capable agent for a task.
Enable auction-based routing by setting the environment variable GARDEN_QUOTA_ROUTING to auction in your worker configuration. This activates the deterministic, CAS-safe award logic for job assignment across your worker fleet.
Deterministic auctioning uses Thompson sampling to resolve job awards without a central auctioneer, ensuring reproducibility across all hosts. It implements CAS-safe award logic to guarantee consistent job assignment throughout the distributed system.
Use a reputation-based auction system when inefficient centralized job distribution bottlenecks your autonomous agent fleet. It is ideal when new coding tasks are posted and multiple worker agents need to bid based on historical success rates and estimated costs.
DUBSOpenHubOrchestrate hierarchical AI agent fleets for distributed task execution.
LGuguiDispatch tasks to worker-agents and track performance metrics.
HeadyAIOrchestrate distributed HeadyBee agents with phi-scaled quotas and CSL scheduling.
NinatuziSchedule cron jobs with collision prevention and quiet hours gating.
