Doesn't apply to every scenario where agents are useful. Think about businesses - most business processes are not optimized this way. They are mostly a conglomeration of hierarchy, relationships, customs, IP, etc. Model development will never cross such paths. Thus, harnesses are needed. A foundational system that covers 80% of business processes has value, and I'm targeting that problem --- https://github.com/rush86999/atom

Counter point to “complex businesses” is they’re currently buying Copilot licenses for their employees, with the hopes it can intermingle with Office365. Very ineffective, I think, buy from my exposure the “non-tech” side of things is not on the same timeline as us.

That’s also one of the reasons why I think harnesses might be obsolete in a year, or re-defined completely.

Obviously, once again, I might be completely wrong. But my thesis is investing hard on any model/harness/custom builds might be counter productive given the speed of development. It’s not easy, but having a nimble org where you can swap things around might be a good call.

It's like everyone knows the dirty truth, but no one wants to do anything about it.

Why? Self-interest has made everyone blind to open corruption.

The worst part is reality hits 3-4 years later, and the big guy is already out and about. Already with his billions, partying in the Caymans.

Welcome to democratic oligarchy!!

The worse part is whatever competent hoodlums will arrive after him, to enjoy the gates he has crashed

Here's a better summary: https://github.com/rush86999/atom/blob/main/docs/architectur...

I also have a self-evolving architecture on it, but you have to be careful, especially if you're using multiple models (still WIP).

That looks like a “Deep Research” report generated by an LLM, not a summary.

Yeah, it is. Bad wording on my part, but the list is a summary of what's going on so far in agent harnesses.

Luckily, I have the coding plan for z.ai, so I'm happy with this model as I always kept running out of usage with the original glm-5.3

I think you've really hit the mark on how the harness should be structured:

1. Guardrails - deterministic, social intelligence, team alignment & accountability 2. Learn by doing 3. make it stupid easy for the agent to research and access data 4. DRY

Research supports this. Try picking up some ideas from my harness: https://github.com/rush86999/atom

Hi. This is very interesting, could you link to the research? There is a dearth of proper research studies that A/B test what approach is best in terms of harness structure based on repeatable benchmark data with relevant sample uses-cases.

Reasoning / self-consistency (voter in core/llm/self_consistency_voter.py): - Wang et al. Self-Consistency Improves Chain-of-Thought — ICLR 2023, Google Brain, 4k+ cites — https://arxiv.org/abs/2203.11171 — N-sample majority vote we use verbatim - Chen et al. Universal Self-Consistency — ICML 2024 — https://arxiv.org/abs/2311.17311 — judge fallback when no hash collides - Soft Self-Consistency — ACL 2024 — https://aclanthology.org/2024.acl-short.28.pdf - Too Consistent to Detect — EMNLP 2025 — https://aclanthology.org/2025.emnlp-main.238/ — why SC doesn't fix systematic bias - Self-Consistency Falls Short — TACL — https://direct.mit.org/tacl/article/doi/10.1162/TACL.a.625/ — position-bias failure mode

Multi-agent / org (core/agent_radio/, core/fleet_orchestration/): - Stanford Virtual Biotech — bioRxiv 2026.02.23.707551, Zou Lab — https://www.biorxiv.org/content/10.64898/2026.02.23.707551v1 — 37k agents, CSO->scientists->reviewer->re-delegation, Merck external validation of B7-H3 design. Basis for VFS + hierarchy. - Debate or Vote (Choi & Li) — NeurIPS 2025 — https://arxiv.org/abs/2508.17536 — MAD gains = majority vote, not debate (why we didn't build debate)

Sandbox / eval: - DABstep — arXiv:2506.23719 — https://arxiv.org/abs/2506.23719 — 450 real Adyen tasks, justifies code-interpreter + sandbox isolation - Spotlighting — Microsoft Research — https://arxiv.org/abs/2403.14720 — provenance delimiters cut injection ASR 50% -> <2%

Governance: - OWASP Top 10 for Agentic Applications 2026 — globally peer-reviewed by 100+ experts, Dec 2025 — https://genai.owasp.org/resource/owasp-top-10-for-agentic-ap... — HIGH. Atom maps 1:1 (Goal Hijack -> match-confidence + oracle, Tool Misuse -> sandbox whitelist/caps, Privilege Abuse -> capability bindings, Memory Poisoning -> verified-episode graduation, etc.) docs/marketing/RESEARCH_NOTES.md:130 - NIST AI Agent Standards Initiative — Feb 17 2026, NIST CAISI — https://www.nist.gov/artificial-intelligence/ai-agent-standa... + RFI summary May 2026 https://www.nist.gov/publications/summary-analysis-responses... — HIGH (US gov standard). Defines the 4 enterprise minimums Atom implements: identification, authorization, access delegation, logging. - Stanford Virtual Biotech — bioRxiv 2026.02.23.707551 — https://www.biorxiv.org/content/10.64898/2026.02.23.707551v1 — CSO -> 4 divisions -> 8 scientists -> reviewer -> re-delegation, no debate, no SFT — HIGH (Stanford Zou lab + Merck external validation). Basis for Atom's fleet hierarchy core/agent_radio/ and why maturity is routing not security. - Spotlighting — Microsoft Research — https://arxiv.org/abs/2403.14720 — HIGH — provenance delimiters <provenance type="tool_output"> cut indirect injection ASR 50% -> <2%, used in core/provenance.py:10 - IntentGuard — https://arxiv.org/abs/2512.00966 + OpenReview — HIGH — intent tracing ASR 100% -> 8.5% on AgentDojo/Mind2Web, basis for sandbox egress allowlist + core/sandbox_tripwire.py

Mine is a WIP for automation but has similar concepts to what you're looking for: https://github.com/rush86999/atom

AI agent workforce: https://github.com/rush86999/atom

Now with OpenCode Go integration. I'm rapidly fixing as many bugs as I can, and with the same subscription. Thank you, DeepSeek!

It's opus 5 > Kimi K3 > grok 4.5

That's a pretty good benchmark

I'm still working on mine: https://github.com/rush86999/atom/blob/main/docs/architectur...

Cheers, good luck!