Harness Autonomous Worker Agents · Harness
Launched GA June 30 2026, these are pipeline-native reasoning agents for the software-delivery stage (between code and production) rather than code authoring: every pipeline step can run as an agent governed by existing enterprise deployment controls, with an LLM Gateway enforcing policy on prompts, distinct auditable AI identities, per-agent/per-pipeline model switching, and a marketplace of Harness-managed and community agents.
Identity
| Homepage | https://www.harness.io/press-and-news/harness-launches-autonomous-worker-agents-for-software-delivery |
|---|---|
| License | proprietary |
| Domain | enterprise-workflow |
| Form factor | ci-step · cloud-service · web-app |
Model support
| Model-agnostic | ✓ yes |
|---|---|
| Native models | Claude (direct and via AWS Bedrock) · OpenAI GPT |
| Local models | ? unknown |
| Pricing | enterprise |
Extensibility
| MCP client | ? unknown |
|---|---|
| Agent Skills support | ? unknown |
| Plugins / hooks | ✓ yes |
Execution & safety
| Sandboxing | cloud-remote |
|---|---|
| Autonomy spectrum | background-cloud |
| Subagents | ? unknown |
| Parallel sessions | ? unknown |
| Context management | none-observed |
Integration & evidence
| Headless / API | ✓ yes |
|---|---|
| GitHub integration | ✓ yes |
| Sources |
✓ profile last verified 2026-07-08 · benchmark figures are vendor-reported unless a source states otherwise