Skip to main content
An evaluation arm pairs an agent harness with a model. The harness reads the instruction, calls tools, and manages the conversation. The model supplies its decisions.

Built-in harnesses

Use the model and effort reference to copy the exact strings for each harness. Model names are not interchangeable across harnesses.
gemini is retired. Select antigravity for new runs; historical Gemini CLI records remain readable.
Use -a name@version to pin a harness release. Without a version, Evolve resolves one at job creation and uses that version across the job. The trial records it in agent_info.version.

Configure an arm

An effort or configuration the harness cannot honor is refused when the job is created. Presets and native config are supported on Claude Code and Codex. The platform applies routing settings over your config, and preset settings take precedence where they overlap. Routing, billing, credential, and environment overrides are rejected. The model table shows defaults and reasoning options for each model. Every arm needs an explicit model. The live capability document lists accepted configuration options.
no-internet disables the harness’s server-side web tools. The task’s network policy controls network access from the sandbox.

Register a custom agent

Register an installation source and a command. The name then works with -a.
Use --dir instead of --install-script to upload a directory. The install runs with internet access and no secrets; place executables in $PREFIX/bin.

Execution contract

  1. Install the agent from its registered source.
  2. Prepare the instruction file and gateway environment.
  3. Start run_command with sh -c in the task working directory, with the instruction on stdin.
  4. Route model calls through the supplied gateway credentials.
  5. Collect outputs and run the task verifier.
Your agent must use these gateway settings for metered model calls. If it requires a config file, write that file from these values inside run_command. --agent-env KEY=VALUE adds non-secret runtime settings to the registration. It cannot override the contract variables or declare credential-like names. Attach job secrets for credentials the task needs.

Custom-agent differences

Custom agents do not support non-root task users or task-declared MCP servers. They also have no built-in live trace parser or shared --effort option. Put implementation-specific reasoning flags in the run command. Harness version pins are not accepted for registrations. Registrations belong to a user and an organization. Members can use them; the owner manages them. Deleting one preserves historical job records.

Agent commands

Register, inspect, and remove a custom agent.
Last modified on November 9, 2026