About Replicas
Overview
Replicas runs coding agents such as Claude Code, Codex, Cursor, and Opencode in cloud workspaces. Each task gets its own Linux VM, preloaded with your repositories, tools, and dependencies. Agents install dependencies, run services, drive a browser, and verify their own work inside that machine.
Teams assign engineering work from Slack, Linear, GitHub, GitLab, or the web dashboard. Agents return pull requests, commits, comments, or completed tasks for review. Usage analytics show which source, person, harness, and model produced each run.
Key Benefits
- Each task runs in an isolated Linux VM that shares nothing with other agents.
- Warm hooks preinstall dependencies, images, MCPs, skills, files, and variables before an agent starts.
- Agents drive a real desktop and browser, then hand back screenshots and screen recordings.
- Work starts from existing channels such as Slack and Linear, so teams keep their current workflow.
- Per-minute analytics attribute usage to a source, person, harness, model, and credential.
- Workspaces sleep automatically when idle and wake on demand.
How It Works
Users assign a task from Linear, Slack, GitHub, GitLab, or the dashboard, and Replicas provisions a Linux VM with the configured environment. The chosen coding agent, such as Claude Code or Codex, works on the task inside that VM and can run tests, drive a browser, and record its work. The agent returns a pull request, commit, comment, or completed task, which the user reviews, merges, or sends back for another pass.
Use Cases
- Engineering teams hand code review, recurring maintenance, and feature work to cloud agents.
- Engineers turn Linear issues into agent runs that read the ticket and start implementation.
- Platform teams fire scheduled or webhook-triggered runs that report back without anyone present.
- Developers who already use Claude Code or Codex run the same agents in a cloud environment.
- Engineering leaders measure agent output and attribute usage to people, harnesses, and models.
- Teams mentioning a bug in a Slack channel can start a fix without leaving the conversation.
Why Choose This Product
Replicas gives each task its own Linux VM and supports four coding agent harnesses, so teams can switch harnesses without changing their workflow. The Developer plan caps manual use at 150 hours per month for one individual. Full and Flex seats require the Team plan.
Replicas Pros & Cons
- 14-day trial with no card required
- Isolated Linux VM for every task
- Supports Claude Code, Codex, Cursor, and Opencode
- Delegation from Slack, Linear, GitHub, and GitLab
- Per-minute usage analytics by source, person, and model
- Developer plan has a monthly manual-use limit of 150 hours
- Full and Flex seats are available on the Team plan only
Key Features
Isolated Linux VMs
Each task boots its own Linux VM in the cloud, sharing nothing with the agents running beside it.
Configured Environments
Developer+Warm hooks preinstall dependencies, images, MCPs, skills, files, and variables before an agent starts.
Multi-Agent Harnesses
Runs Claude Code, Codex, Cursor, or Opencode, and lets you mix agents across tasks or inside one workspace.
Delegation From Anywhere
Starts agent runs from Slack, Linear, GitHub, GitLab, or the web dashboard.
Automations and API
Developer+Fires runs from a schedule, a webhook, or one API call, and Replicas provisions the workspace without anyone present.
Full Computer Use
Agents drive a real desktop and browser, then return screenshots and screen recordings of their work.
Usage Analytics
Attributes every workspace minute to a source, person, harness, model, credential, skill, and MCP server.
Reviewable Outputs
Agents return a pull request, commit, comment, or completed task for review, merge, or resend.
Idle Auto-Sleep
Workspaces sleep automatically when idle and wake on demand.
Replicas Pricing
View full pricing →- For one individual, billed upfront
- 150 hours per month
- API and automations billed by usage
- Unlimited repositories, environments, and automations
- Warm pools and warm hooks
- Per Full seat, billed upfront
- Unlimited manual workspace minutes
- API and automations billed by usage
- Warm pools and warm hooks
- Higher API rate limits
- Auto-upgraded sandbox resources (4 vCPU, 32 GB disk, 16 GB memory)
- Pay as you go at $0.016 per minute
- Flex seats pay for manual usage
- Available on the Team plan
- Unlimited manual and automated workspace minutes
- Custom warm hooks and pools
- Custom API rates and rate limits
- SOC 2 and Data Processing Agreement
- SCIM user provisioning
| Feature | Developer $50/mo | Full $200/mo | Flex $0.016 usage-based | Custom Custom |
|---|---|---|---|---|
| For one individual, billed upfront | ||||
| 150 hours per month | ||||
| API and automations billed by usage | ||||
| Unlimited repositories, environments, and automations | ||||
| Warm pools and warm hooks | ||||
| Per Full seat, billed upfront | ||||
| Unlimited manual workspace minutes | ||||
| Higher API rate limits | ||||
| Auto-upgraded sandbox resources (4 vCPU, 32 GB disk, 16 GB memory) | ||||
| Pay as you go at $0.016 per minute | ||||
| Flex seats pay for manual usage | ||||
| Available on the Team plan | ||||
| Unlimited manual and automated workspace minutes | ||||
| Custom warm hooks and pools | ||||
| Custom API rates and rate limits | ||||
| SOC 2 and Data Processing Agreement | ||||
| SCIM user provisioning |
Pricing extracted from the product website and may change. Check the source for current details.
Frequently asked questions about Replicas
Which coding agents can I run?
Claude Code, Codex, Opencode, and Cursor. Every agent runs on the same environment primitives, so you can mix agents across tasks or run several inside one workspace.
What does an agent's environment look like?
Each agent works in an isolated cloud environment with your repository, tools, dependencies, and services ready to use.
Is the output always a pull request?
No. Agents can return a pull request, commit, comment, or completed task for you to review, merge, or send back.
What repositories can I use?
Any GitHub or GitLab repository you have access to. Connect your GitHub or GitLab account and select which repos to enable.
How does billing work?
Usage is measured in minutes of workspace runtime. Developer is for one individual with a monthly manual-use limit. Team Full seats have unlimited manual runtime, while Flex seats pay for manual usage. Paid API and automation usage is uncapped and billed separately. Workspaces automatically sleep when idle and wake on demand.
Is Replicas free?
Replicas is a paid product. See the pricing section for current plans and starting prices.