Skip to content

Awesome list for AI agent harness engineering: tools, patterns, evals, memory, MCP, permissions, observability, and orchestration.

Add bernstein: deterministic audit-grade orchestrator for 40+ CLI coding agents with HMAC-signed audit chains

ai-boost5a83d0bd5315
diff --git a/README.md b/README.md
index f834987..49fe15d 100644
--- a/README.md
+++ b/README.md
@@ -254,6 +254,7 @@ Harness components organized by the problem they solve, not by vendor.
- [Hive](https://github.com/aden-hive/hive) — YC-backed production harness that compiles multi-agent objectives into deterministic execution DAGs with state persistence, crash recovery, cost enforcement, and human-in-the-loop oversight. The most complete open-source reference for running agent workloads in production rather than demos. ![Stars](https://img.shields.io/github/stars/aden-hive/hive?style=flat-square&label=★&color=yellow)
- [thClaws](https://github.com/thClaws/thClaws) — Native-Rust agent harness platform with four surfaces (GUI, CLI, web, one-shot), multi-provider routing, and sovereign-by-design architecture. The most complete open-source reference for a locally-run, offline-capable harness that unifies MCP, skills, AGENTS.md, hooks, and session resumption in a single binary. ![Stars](https://img.shields.io/github/stars/thClaws/thClaws?style=flat-square&label=★&color=yellow)
- [sandcastle](https://github.com/mattpocock/sandcastle) — TypeScript-native orchestration for sandboxed coding agents that treats provider-agnostic isolation (Docker, Podman, or Vercel Firecracker microVMs) as a primitive, not a framework. The built-in review-pipeline and parallel-AFK-agent patterns demonstrate how lightweight harness layers can enforce safety without the complexity of full graph orchestration. ![Stars](https://img.shields.io/github/stars/mattpocock/sandcastle?style=flat-square&label=★&color=yellow)
+- [bernstein](https://github.com/sipyourdrink-ltd/bernstein) — Deterministic scheduler for 40+ CLI coding agents running in parallel git worktrees with an HMAC-signed audit chain, signed agent cards, and per-artefact lineage. The zero-LLM coordination loop and tamper-evident audit trail make it the only open-source orchestrator designed for compliance-sensitive agent fleets. ![Stars](https://img.shields.io/github/stars/sipyourdrink-ltd/bernstein?style=flat-square&label=★&color=yellow)
### Verification & CI Integration