← Back to the Hub
Included
RLM
rlm
Built-in recursive coding harness: one persistent Python tool, child agents, and a kernel that runs on the host, in Docker, or under Pydantic Monty.
rlmOpen sourceReadyContext replay sessions
Host Python permissions: RLM Python runs with the permissions of the SuperQode process, with no approval prompts or sandbox.
Specification
| Category | SuperQode harnesses |
|---|---|
| Runtime | rlm |
| Interface | TUI and CLI |
| Source | built-in |
| Licence | Apache-2.0 |
| Continuity | Context replay |
| Support | Ready |
| Based on | rlm |
Runtime behaviour
Capabilities
single workflowRecursive child agentsStructured run events
Tools
pythonPolicies and ownership
- +Sandbox: none
- +Approvals: none
- +Shell: allowed
- +Network: allowed
- +Checks: not required
Get started
Inside SuperQode
$ :harness switch rlm$ :harness show rlm$ :harness statusFrom your shell
$ superqode harness run rlm "Describe your task"$ superqode harness show rlmEvaluate and optimize
Measure
$ superqode harness test --spec harness.yaml$ superqode harness eval --spec harness.yaml --tasks eval-tasks.yaml$ superqode harness bench --manifest harnessbench.yamlImprove
$ superqode harness optimize-omni --spec harness.yaml --tasks eval-tasks.yaml$ superqode harness optimize --spec harness.yaml --tasks eval-tasks.yaml$ superqode harness promote stageALSO IN SUPERQODE HARNESSES
Browse all 104 harnesses