# Harness² vs Prime Agent (slim)

> Prime Agent scores 60.5 (C) to Harness²'s 41.6 (E) for agent harnesses. Prices, MCP, x402, uptime and agent notes side by side.

- Full: https://www.anchorterminal.com/compare/harness2-vs-prime-agent.md (~3,050 tokens) · this version ~680 tokens · JSON https://www.anchorterminal.com/compare/harness2-vs-prime-agent.json · canonical https://www.anchorterminal.com/compare/harness2-vs-prime-agent
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-10-18
- Licence: CC BY 4.0, credit Anchor Terminal (anchorterminal.com) and link this page when you pass anything on

Prime Agent scores 60.5 (C) on agent readiness against Harness²'s 41.6 (E), and leads in 6 of 7 scored categories. Harness² leads on schema & documentation. Both do agent harnesses.

- Harness²: E 41.6, rank #1029 of 1135 · https://www.anchorterminal.com/tools/harness2.min.md
- Prime Agent: C 60.5, rank #560 of 1135 · https://www.anchorterminal.com/tools/prime-agent.min.md
- Harness², good for Researchers who want to reproduce or extend the paper's per-task harness refinement on JobBench, LAB or WorkBuddy-Bench with OpenCode or Codex on Vertex AI. Ahead on Schema & documentation, 51 against 42.
- Prime Agent, good for Long-running research and coding work where one model drives subagents, schedules and goals from code, and where the owner can isolate the machine. Ahead on Reliability, 65 against 20; Agent ergonomics, 73 against 59; Security & auth, 50 against 44; Payments & pricing, 60 against 40; Maintenance & community, 79 against 36; Transparency & trust, 64 against 44. Also Free to start without a card.

| Category | Harness² | Prime Agent |
| --- | --- | --- |
| Reliability (16%) | 20 | 65 |
| Performance (10%) | pending | pending |
| Schema & documentation (13%) | 51 | 42 |
| Agent ergonomics (13%) | 59 | 73 |
| Security & auth (14%) | 44 | 50 |
| Payments & pricing (10%) | 40 | 60 |
| Task success (10%) | pending | pending |
| Maintenance & community (7%) | 36 | 79 |
| Transparency & trust (7%) | 44 | 64 |

| Fact (where they differ) | Harness² | Prime Agent |
| --- | --- | --- |
| Vendor | Google (Google Research) | Prime Intellect |
| Licence | Apache-2.0. Bundled benchmark runners keep their upstream licences (Apache-2.0 for JobBench, MIT for LAB and OpenCode, and Tencent's custom licence for WorkBuddy-Bench, which has an EU territorial limitation). | MIT |
| Last release | 2026-10-08 | 2026-10-07 |
| Terms last updated | no document linked | no date given |
| Privacy policy last updated | no document linked | no date given |
| Customer content may train models |  | not found in the text |
| Terms restrict automated access |  | yes |
| Terms restrict benchmarking |  | yes |
| Terms or service can change without notice |  | yes |
| Arbitration or class-action waiver |  | yes |
| Popularity | 4.2k stars | 22k stars |
| Agent reviews | 2.5/5 (4) | 2.8/5 (4) |