Isorun
$19.36 a month*
Home / AI agent sandboxes / DigitalOcean Agent Harness Runtime
Hardware-isolated microVM sandboxes for agents and code, billed per second on CPU used and memory allocated.
Verdict (2026-11-13): DigitalOcean Agent Harness Runtime scores 61/100, #38 of 63 AI agent sandboxes. Strong on isolation, persistence, docs and SDKs. Weak on price. Best for: Running hosted coding agents such as Claude Code, Codex or OpenCode in isolated machines.
Skip it if: Public preview, and only one region (RIC1) is listed.
Prices checked 2026-11-15.
Harness Runtime is part of DigitalOcean Managed Agents, in public preview for all users. Each session runs in a hardware-isolated microVM that can host a coding agent such as Claude Code or Codex, or act as a plain sandbox for running commands. CPU is billed on actual consumption and memory on allocation, per second, and paused sessions stop compute charges while keeping files and history. Checkpoints capture the whole machine so you can fork or roll back.
61/100 in AI agent sandboxes, rank 38 of 63. Parts with no data are left out and the rest is rescaled.
Full DigitalOcean Agent Harness Runtime pricing: plans and costs at four workloads →
$19.36 a month*
$11.73 a month*
$10.14 a month*
$19.00 a month*
$17.95 a month*
$28.44 a month*
$20.00 a month*
$20.00 a month*
DigitalOcean Agent Harness Runtime is not in the public benchmarks we track yet. See all benchmarks.
| DigitalOcean Agent Harness Runtime | |
|---|---|
| Category | agent sandbox |
| Our score | 61/100 |
| Isolation | Hardware-isolated microVM per session |
| vCPU per hour | $0.044 (active CPU only) |
| Memory per GiB-hour | $0.0095 |
| Storage per GiB-month | $0.05 |
| Billing | per second |
| Free tier | New customers receive a one-time $5 promotional credit for Managed Agents usage. |
| Pause and resume | Yes |
| Snapshots | Yes |
| GPU | No |
| Custom images | Yes |
| Network controls | Yes |
| Max vCPU | 16 vCPU |
| Max memory | 32 GiB |
| Open source | No |
| Self-host | No |
| SDKs and interfaces | doctl CLI, REST API, PyDo (Python), Control panel, Cuttlefish iOS app |
| Regions | RIC1 (Richmond) only, per the availability table. |
| Facts checked | 2026-11-13 |
No reviews yet. Have you used DigitalOcean Agent Harness Runtime? Your review helps other developers choose.
DigitalOcean Agent Harness Runtime scores 61/100, #38 of 63 AI agent sandboxes. Strong on isolation, persistence, docs and SDKs. Weak on price. Best for: Running hosted coding agents such as Claude Code, Codex or OpenCode in isolated machines. Facts checked 2026-11-13.
$50.00 a month for 2 vCPU, 4 GiB, 176 hours a month (8 h × 22 days), CPU busy 30%. Free tier: New customers receive a one-time $5 promotional credit for Managed Agents usage.
Hardware-isolated microVM per session.
In AI agent sandboxes, the top-rated are Isorun (78/100), Northflank (78/100), Koyeb Sandboxes (76/100). DigitalOcean Agent Harness Runtime is #38 of 63.
| CPU | $0.044 per vCPU-hour | Actual CPU consumed |
| Memory | $0.0095 per GB-hour | Actual memory allocated over time |
Snapshot and checkpoint storage is billed at $0.05 per GiB-month for the data you store.
| Pro | $50 | 15% | $58.82 |
| Team | $200 | 20% | $250.00 |
Outbound traffic from the sandbox to the public internet is billed at $0.01 per GiB of egress data.
Pausing a session stops compute charges while preserving the sandbox’s files and conversation history
New customers receive a $5 promotional credit for Managed Agents usage.
| XLarge | `mars-16vcpu-32gb` | 16 | 32 GB | 500 GB | 512 Mbps |
You can also create checkpoints to capture the whole machine state from which you can fork a new session or roll back to a prior state.
Sandbox egress is unrestricted unless the environment spec sets an allowlist.
Each session has a monthly active compute cap of 744 hours.
Managed Agents, available in public preview for all users, bring together hardware-isolated microVM sandboxes
Harness Runtime is available in the regions shown below.