multi-account-gatewaylisted

Give an always-on agent enough model capacity that a long overnight run never dies halfway from a hidden limit — the ToS-clean way. Explains the recommended budgeted API-key lane (a spending cap plus an automatic fallback to a second provider), and the wrapper-with-fallback pattern (provider-first, direct-CLI fallback), 429 cooldown, and token refresh in plain terms. Use when a scheduled or overnight run keeps stopping on a rate limit or quota wall, when you're on a subscription plan and worried about capacity, when you're tempted to "stack accounts" to get more headroom, or when you want to know the one capacity trick that gets accounts banned so you never do it by accident.
search-atlas-group/amm-founding-circle · ★ 0 · AI & Automation · score 72

Install: claude install-skill search-atlas-group/amm-founding-circle

# multi-account-gateway **This is a guide, not a tool.** It answers one question every agency owner hits the first time they leave an agent running unattended: *"How do I make sure a long job doesn't quietly run out of capacity at 3am — without doing something that gets my account banned?"* There is a right answer and a wrong answer, and the wrong answer is easy to stumble into because it looks clever. This guide leads with the right answer, explains the mechanics of how "keep working when one source is busy" actually works, and then draws a hard red line around the one pattern that gets people banned — so you never cross it by accident. > This skill is the **fuel line** of always-on. The `host-your-agent` skill gets your agent off your laptop and onto a schedule; the `night-shift` skill is the contract that bounds every unattended run (time box, read-only-by-default, a failure ledger). This guide is the piece that keeps the tank from running dry mid-run. Read those two first — this one assumes you already have a scheduled run you're trying to keep alive. --- ## The one-minute version - **The most common way an unattended run dies is quietly running out of model capacity.** It's 3am, the limit is hit, the run stops, and you find out at noon. - **The recommended fix is a budgeted API key with a spending cap you set, plus an automatic fallback to a second provider if the first is down.** Predictable, yours, and a cap means a runaway job *stops* instead of surprising you w