switchboard
U

Unreal Agent

by Unreal Labs

Runs coding agents on an async-first Go harness with persisted, forkable sessions, as a library or the unreal-agent-runner CLI, for developers building or benchmarking agents.

4
Skills
None
Auth
No
Streaming
No
Push

Skills

Agent Runner CLI

Runs an agent from a prompt or JSON request in a chosen workspace, writing events to stdout as JSONL and exiting when the task finishes.

Multi-Provider LLM Support

Switches the LLM backend between openai, openai-codex, openrouter, fireworks, and ollama with environment variables for provider and model.

Forkable Sessions

Persists append-only session history to disk so runs can be resumed or forked, with versioned, serializable session items.

Harbor Benchmark Adapter

Evaluates unreal-agent-runner with Harbor on tasks such as Terminal-Bench on Docker or Modal, producing ATIF trajectories and logs.

Code & DevToolsAutonomous Agentsagent-harnessgolangasynccoding-agentharbor-benchmarksopenrouterollamaopen-source
Visit Agent
unreal-agent
Runs coding agents on an async-first Go harness with persisted, forkable sessions, as a library or the unreal-agent-runner CLI, for developers building or benchmarking agents.
fields
nameUnreal Agent
providerUnreal Labs
urlhttps://github.com/unreallabsai/unreal-agent
categoriescode-devtools · autonomous-agents
accessapi · cli
authnone
streamingfalse
pushfalse
verifiedtrue
tagsagent-harness, golang, async, coding-agent, harbor-benchmarks, openrouter, ollama, open-source
skills
runner-cliAgent Runner CLIRuns an agent from a prompt or JSON request in a chosen…
multi-providerMulti-Provider LLM SupportSwitches the LLM backend between openai, openai-codex, …
forkable-sessionsForkable SessionsPersists append-only session history to disk so runs ca…
harbor-benchmarksHarbor Benchmark AdapterEvaluates unreal-agent-runner with Harbor on tasks such…