AIDevToolsProductivity

QuotaPilot – agent quota router

Route coding-agent tasks to cheap sub-agents by difficulty and stop burning quota

Problem
Heavy coding-agent users on personal plans complain that quotas drain absurdly fast after switching models, and they can't tell which tasks deserve the expensive flagship. Some already hand-route simple jobs to cheap sub-agents, but there is no lightweight routing layer in their workflow.
Solution
A local proxy/MCP layer that intercepts every request you send to the flagship model and rewrites routing by task type. Simple jobs (boilerplate, tests, scripts) go to cheap models like DeepSeek; hard reasoning stays with the flagship. A dashboard shows quota saved per request, weekly burn curve, and sub-agent hit rate.
Users
Indie developers and solo builders paying for Codex or Claude Code out of pocket
MVP
Local proxy for CLI traffic; built-in task classification rules; one-click 'cheap vs faithful' policy switch; daily quota saved report; exportable shareable configs
Revenue
Free basic routing plus paid policy packs and team-shared configs, subscription at $5-9/month
Why now
Multi-model and sub-agent orchestration just became mainstream, while token budget pressure turned routing from a nice-to-have into a necessity.

Source discussions · 4

  1. Hacker NewsShow HN: Flash-Agents – DSH as MCP for Claude4↑ 0
  2. Hacker NewsAsk HN: How are AI budgets changing in your company?2↑ 1
  3. V2EX[问与答] 怎么让 codex 消耗额度慢一点?0↑ 5
  4. V2EX[问与答] 有没有同时使用 5.6 sol high 和 6.1 sol high 的 plus 用户,想咨询个问题
← Back to list