Ops & admin automation
Hand repeat admin (receipts, logs, notices, handoffs) to workflows your agent can trigger.
Independent reviews. IonBin has no partnerships with the tools on this page and was not paid for these listings. Our evaluations are our own, based on public documentation and our research. Product names are trademarks of their owners and are used for identification only. How disclosure works.
Concept
Stack 06 Stack 08
Background agents
Verified Oct 17, 2026 Review by Apr 15, 2027 3 sources
Short answer
A background agent runs on the vendor's servers rather than your laptop. You give it a goal, a schedule or a trigger, and it works while you are away, then reports back for review. Claude Code routines, Google's Gemini Spark and scheduled tasks in Google Antigravity work this way. The trade-off: the vendor holds the credentials the agent acts with.
What it is
Most AI work starts when you type. A background agent starts on its own: on a schedule, from an API call or when an event fires, such as a new GitHub pull request. It keeps working after you close the laptop.
How it works
- Claude Code routines (research preview, checked 2026-10-17) run on Anthropic’s cloud. A routine starts on a schedule (hourly, nightly, weekly or once), from an API call or on GitHub events.
- Gemini Spark, announced on May 19, 2026, is a 24/7 agent built on Gemini 3.5 and the Antigravity harness. Google says it asks first before high-stakes actions such as spending money or sending email. It went to trusted testers first, then a Beta for Google AI Ultra subscribers in the US.
- Google Antigravity added subagents, hooks, scheduled tasks and asynchronous task management at I/O 2026.
Benefits
- Work you no longer start by hand. Weekly reports, dependency updates and inbox digests run on time.
- Parallel work. Several agents can take separate tasks at once.
- A natural review rhythm. You check finished output instead of watching each step.
Limits and risks
- Credentials live with the vendor. The agent can only do what its connectors allow, so scope each one tightly. See letting an agent act safely.
- Unattended mistakes compound. Put a human approval step before anything that sends, pays, publishes or deletes.
- Preview features change. Several of these are previews or limited to certain plans and countries. Check the vendor’s docs before you rely on one.
Checked specs
Snapshot from official sources on Oct 17, 2026.
| Claude Code routines (checked 2026-10-17) | Research preview; run on Anthropic's cloud; triggers are a schedule (hourly, nightly, weekly or once), an API call with a bearer token, or GitHub events [1] |
| Gemini Spark | Announced May 19, 2026: a 24/7 agent on Gemini 3.5 using the Antigravity harness; asks first before high-stakes actions such as spending money or sending email [2] |
| Gemini Spark rollout | Trusted testers first, then a Beta for Google AI Ultra subscribers in the US [2] |
| Google Antigravity (I/O 2026) | Added subagents, hooks, scheduled tasks and asynchronous task management [3] |
Best-fit stacks
Dev, web & IT
Design, code, host and ship sites and internal tools, with an AI editor doing most of the typing.
Agent compatibility
| Tool in this entry | How agents connect |
|---|---|
| Claude | Agent platform API/CLIBulk import |
| Gemini (Google) | Agent platform API/CLIBulk import |
| Google Antigravity | Agent platform API/CLI |
| GitHub | MCPAPI/CLI |
Levels come from each tool's recorded, sourced compatibility. How the levels work.
Related entries
Sources
- Automate work with routines · Anthropicofficial accessed Oct 17, 2026
- The Gemini app becomes more agentic, delivering proactive, 24/7 help · Google, 2026-05-19official accessed Oct 17, 2026
- Subagents, hooks, scheduled tasks and more (I/O 2026) · Google Antigravityofficial accessed Oct 17, 2026
Practitioner views are restated in our words and labeled as views, not measured results. Product facts come from official sources dated above.
Suggest an edit
IonBin's wiki isn't user-edited. Corrections go to our Directory Curator, who checks them against primary sources before anything changes. Spot something wrong or out of date on “Background agents”? Email us with a link to a primary source.
Suggest an edit by email or write to [email protected]