Claude Code Max subscriber details multi-host fleet pain points in GitHub issue, mulling switch
jyongchul · ghdev · 2026-10-18
A Max subscriber running Claude Code as an orchestrator across one Windows workstation and four headless Linux servers (dozens of tmux agent sessions) filed issue #103179 on anthropics/claude-code listing pain points after months of daily use:
- Manual multi-host login: each server requires a separate OAuth code paste; he asks for a documented device flow or org-managed token to authorize a whole fleet at once
- Over-refusal: as CEO, asking the agent to place his own company's seal on his own documents (routine Korean e-seal practice) was blocked
- Agent hands work back: frequent "please do X yourself" stops despite explicit standing autonomy instructions
- Usage limits drain fast for agent fleets, making cost hard to justify vs cheaper alternatives
More from coding & agent
- Claude Code Pushes Urgent Security Update Across All Installations — umeshonai · 2026-10-23
- Claude cut a 2.6-hour video export to 20 minutes — but only after being told to rebuild the server — byhewar · 2026-10-23
- 3 flight-booking rules my AI agent checks that I'd never do by hand — Efistoffeles · 2026-10-23
- OpenJet claims Qwen3.8-27B coding runs 1.9x faster than llama.cpp on one RTX 3090 — JeremyCMorgan · 2026-10-23
- Building an autonomous SEO agent in Hermes with three skills: programmatic pages, Keywords Everywhere, DataForSEO — EXM7777 · 2026-10-23
- Building a sports betting AI agent with onchain execution and human approval by default: architecture and lessons — Dry-Comfortable-2678 · 2026-10-23