AI Productivity

Fallback Chains

Give any type of work an ordered list of backup agents, and pick which failures hand off to them — launch error, provider unavailable, rate limit, quota exhausted, or no response. A run that would have died on one provider's limit continues on the next engine instead.

Screenshot coming soon

Why You Need This

How this feature boosts your daily productivity

A long agent run is only as reliable as the provider behind it. Hit a rate limit at 2am and the old outcome was the same every time: the run stopped, nothing came back, and you found out in the morning and started again somewhere else. Fallback chains put that recovery in the table rather than in your head. In Defaults — the Router, renamed and moved to Advanced because it is consulted at exactly two moments, unnamed work and a failed run — every type of work can carry up to four ordered backups. You choose which failures count as a trigger, so a genuine launch error and a quota that ran out can be treated differently, and you choose whether the chain advances on its own or waits for your approval before spending someone else's quota. When a hop happens it is written down rather than hidden: an active team shows the line that explains its own output — Codex rate-limited → Claude — so a result that came from the backup never reads as if it came from your first choice. Naming an agent still wins over all of it; Defaults only speaks up for work you never assigned, and for a run that has already failed.

How It Works

Get started in 4 simple steps

1
1

Open Orchestration → Advanced → Defaults

The old Router, renamed. It is consulted at exactly two moments: work you never named an agent for, and a run that failed. On first launch its table is seeded from the AI CLIs you actually have installed.

2
2

Give a work type an ordered list of backups

Pick a work type — implement, review, research — and stack up to four fallback steps behind its primary agent, each with its own agent, model and terminal or headless substrate.

3
3

Choose which failures trigger the chain

Launch error, provider unavailable, rate limit, quota exhausted and no response are separate triggers, so a genuine crash and an exhausted quota can be handled differently.

4
4

Decide whether it advances on its own

Let the chain hop automatically, or hold it so a run waits for your approval before it starts spending another provider’s quota.

5
5

Read the hop in the running team

When a fallback fires, the active team shows the line that explains its own output — Codex rate-limited → Claude — so a result from the backup never reads as one from your first choice.

Without 1DevTool

The manual way vs. the 1DevTool way

The Hard Way

# An agent hits its provider rate limit mid-run
# The run ends — it does not pause
# Nothing comes back, and nothing says why
# Find out hours later from an empty log
# Rebuild the run by hand on another engine

The 1DevTool Way

Every work type carries an ordered list of backups.
You choose which failures trigger the hop.
Auto-advance, or hold the run for your approval.
The team records it: Codex rate-limited → Claude.
A named agent still wins over Defaults, always.

Key Capabilities

Everything included with Fallback Chains

Stack up to four ordered backup agents behind any type of work
Trigger on launch error, provider unavailable, rate limit, quota exhausted or no response
Give each fallback step its own agent, model and terminal or headless substrate
Advance the chain automatically, or hold it for your approval
Read the hop that actually happened — Codex rate-limited → Claude
Start from a Defaults table seeded with the CLIs you really have installed
Get a suggestion when one work type keeps failing on one engine
Keep a named agent winning over Defaults, always

Why It Matters

The productivity impact of having ai productivity built into your workspace

Searchable Prompt Library

Every prompt you've ever sent is captured and searchable. Find and reuse your best AI interactions.

Visual AI Communication

Annotate screenshots and embed them in prompts. Show your AI agent exactly what you see.

Session Knowledge Base

AI conversations become searchable history. Find how you solved a similar problem last week.

Power Navigation

Command palette gives you instant access to any feature. Navigate 1DevTool at the speed of thought.

Ready to boost your workflow?

Download 1DevTool for free and experience Fallback Chains along with 170 other features — all in one app.