---
title: "agent-harness"
description: "Use when making a repo agent-ready, verifying harness consistency, checking for documentation drift, bootstrapping harness infrastructure (AGENTS.md as index, docs/ structure, CI verification, enforcement mechanisms), or"
canonical: https://agentpluginsdirectory.com/plugins/agent-harness
last-updated: 2026-09-22
---

# agent-harness
Use when making a repo agent-ready, verifying harness consistency, checking for documentation drift, bootstrapping harness infrastructure (AGENTS.md as index, docs/ structure, CI verification, enforcement mechanisms), or auditing repo agent-readiness maturity level.
- Slug: agent-harness
- Publisher: Netresearch DTT GmbH
- Repository: https://github.com/netresearch/agent-harness-skill
- Manifest: plugin.json
- Version: 1.8.3
- License: (MIT AND CC-BY-SA-4.0)
- Category (editorial): agent-tooling
- Skills: 1 (agent-harness)
- MCP servers: 0
- Stars: 5
- Repository created: 2026-03-22
- Repository last pushed: 2026-09-21
- Publisher type: Organization
- Listing: https://agentpluginsdirectory.com/plugins/agent-harness
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What agent-harness does, in the publisher's words

Agent Skill for bootstrapping, verifying, and enforcing agent-harness infrastructure in repositories. Makes repos agent-ready with self-sustaining enforcement mechanisms that work for all contributors, human or AI, with or without skills installed.

Agent harness is repo-level infrastructure that makes AI coding agents reliable. Instead of relying on individual skill installations, the harness embeds enforcement directly into the project via CI workflows, git hooks, and conventions.

The skill follows the "verify-first" principle: it primarily checks consistency, secondarily creates missing artefacts, and delegates specialised work to existing skills.

- AGENTS.md as compact index (<150 lines), not encyclopedia
- Three enforcement layers: hard (CI/branch protection), automatic (.envrc/hooks), soft (conventions)
- Three maturity levels: Basic, Verified, Enforced
- The skill is the installer; the harness enforces itself

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/netresearch/agent-harness-skill/HEAD/README.md

## Skills

- agent-harness: Use when making a repo agent-ready, verifying harness consistency, checking for documentation drift, bootstrapping harness infrastructure (AGENTS.md as index, docs/ structure, CI verification, enforcement mechanisms), or auditing repo agent-readiness maturity level.

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.