Radar / AI coding tools / Tetriz Harness
Tetriz Harness - open-source tooling for coding-agent fleets
Two open-source tools: Harness Recommendations audits a team's coding-agent rules, skills, hooks, and permissions for breakage and gaps; Harness Control Plane rolls approved fixes out to every machine running Claude Code, Codex, Cursor, or Kiro.
Why it matters
Treats the coding-agent harness as fleet infrastructure - evidence-backed config audits plus one-command team-wide distribution.
What you could build with it
A platform team could audit every engineer's coding-agent rules for breakage and roll out approved configs team-wide with one command.
Does it hold up?
Early positive signals but thin evidence. Both tools are MIT-licensed and self-hosted, and the launch post documents a quick start that scans a sample repo with no model or account required; the team has shipped two prior open tools. No evidence yet of a real team-wide rollout's impact - verdict on effectiveness is pending independent usage.
Built with Tetriz Harness
- Tetriz Harness Reviews (2026) | Product HuntProduct Hunt · User reviews of the launch - early adopter reactions since Oct 15.
- Agent Desk + Session Intelligence | Product HuntProduct Hunt · Tetriz's earlier launch batch (the tools Harness builds on): a Mac menu-bar app showing agents' prompts and harness per project, plus a Go library that evaluates coding sessions turn by turn.
- tetriz-ai/tetriz-harnessgithub · 18 ★ · The launched monorepo: recommends coding-harness changes from repo context and coding signals, then distributes approved rules, skills, hooks, and agents to selected people or teams across Claude Code, Codex, Cursor, and Kiro.
- tetriz-ai/agent-deskgithub · 16 ★ · Companion tool (prior batch): local macOS menu bar app showing AI coding sessions per tool and project, and the skills, rules, agents, and hooks each agent loads.
- tetriz-ai/session-intelligencegithub · 21 ★ · Companion tool (prior batch): Go library that evaluates coding sessions turn by turn - prompt clarity, context, rework, outcomes, wasted effort, plus security flags - self-hosted, bring-your-own-LLM.
Learn more
First spotted on article: source.
More AI coding tools
gstackThe YC CEO's personal Claude Code toolchain: 23 opinionated slash-command tools that turn the agent into a virtual…coding · JEV 0.8Graphify: turn any codebase into a queryable knowledge graph for your coding agentOpen-source CLI plus agent skill that builds a traversable knowledge graph of code, docs, PDFs, and images so…coding · JEV 0.78REA — Reverse Engineer Anything (morluto/rea)MCP server that lets AI coding agents reverse-engineer apps they've never seen the source for — from app behavior down…coding · JEV 0.77OpenCode 2.0 — terminal, desktop and web on one background serviceMajor 2.0 release of the open-source, model-agnostic coding agent: one shared background service powers the terminal…coding · JEV 0.76Orca - open-source agent development environment for a fleet of coding agentsDesktop (plus mobile) app that runs any CLI coding agent side by side, each in its own git worktree, with a board…coding · JEV 0.76mattpocock/skills: Skills for Real EngineersMatt Pocock's open collection of production agent skills, published straight from his .agents directory - 272k stars…coding · JEV 0.75
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.