The governance substrate for AI coding agents — adversarial review, drift-detected rules, immutable audit, closed-loop telemetry.
The agents (Claude, Codex, Gemini, OpenCode) do the engineering. nexus-agents enforces the rules they have to follow, reviews their work adversarially, and audits everything they touch.
Org created 2026-05-17 as the canonical home for the nexus-agents ecosystem. Repos are being migrated from github.com/williamzujkowski/nexus-* in stages; clone URLs auto-redirect during transition.
- nexus-agents — core CLI, MCP server, consensus engine, multi-CLI routing, audit chain
- nexus-toolkit — E2E test suite for nexus-agents MCP tools
- nexus-eval-template — scaffold for new benchmark harnesses (
BenchmarkAdaptercontract) - nexus-eval-* — benchmark harnesses: SWE-bench, SWE-bench Pro, Aider Polyglot, LiveCodeBench, ATBench, TAU-bench
Repo links will be updated here as transfers complete.