Harness Engineering

If you're working with AI coding agents at any serious scale, this is the scaffolding layer you're missing. It builds what the docs call an "agent-first engineering harness": AGENTS.md as a navigation map, domain boundary specs in .harness/*.yml, enforcement rules, and quality scoring. The workflow is depth-first across eight phases, from assessment through verification. The core insight is sound: agents can only reason about what's in context, so you encode architecture, principles, and process patterns directly in the repo instead of leaving them in Slack or team knowledge. Most useful when you notice agents replicating bad patterns or struggling with architectural drift, which usually means your harness is missing or stale.

skills.sh installs: 305
GitHub stars: 222
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add pproenca/dot-skills --skill harness-engineering --agent claude-code

Installs into .claude/skills of the current project.

Files

Harness: Agent-First Engineering Scaffolding

The harness is the scaffolding that makes coding agents effective in a repository. It encodes the knowledge, boundaries, and rules that an agent needs to reason about the full business domain directly from the repo itself.

The philosophy: agents execute, humans steer. The engineer's job is not to write code but to design environments, specify intent, and build feedback loops. The harness is what makes this possible.

Why It Matters

From the agent's point of view, anything it can't access in-context while running effectively doesn't exist. Slack discussions, Google Docs, tacit team knowledge — all invisible. The harness makes this knowledge legible by encoding it as repository-local, versioned artifacts.

A well-harnessed repo gives agents: