agentmods.dev

harness

harness is a plugin for Claude Code from waqas1412/claude-harness. Its manifest loads nothing; the 9 skills, 16 agents, 2 hooks it bundles cost 2,692 tokens per session together, scanned A, original, MIT.

A set of reusable tools for working with Claude Code, an AI coding assistant. It includes read-only advisor agents, task-focused commands, project setup, and checks that enforce repository rules.

In plain words
What is it for?
Use it to set up a repository for the harness, ask for architecture, design, code-quality, performance, or documentation advice, create pull requests or tickets from project profiles, and enforce selected rules with hooks.
Why use it?
It helps keep AI-assisted development consistent by giving the assistant documented project conventions and separate roles for advice, pull requests, tickets, and reviews. Read-only advisors can inspect work without changing files.

Plugin for Claude Code

Written for Claude Code: a Claude Code plugin manifest. Also seen: mentions subagents.

not rated 1 4d ago A scan tokens not measured original MIT
View the source file waqas1412/claude-harness
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Claude Code
/plugin marketplace add waqas1412/claude-harness
agentmods
npx agentmods add plugins/waqas1412/claude-harness/harness
Clone the repo
git clone --depth 1 https://github.com/waqas1412/claude-harness

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for harness

README.md
[![agentmods](https://agentmods.dev/badge/plugins/waqas1412/claude-harness/harness.svg)](https://agentmods.dev/plugins/waqas1412/claude-harness/harness)
Your own site
<a href="https://agentmods.dev/plugins/waqas1412/claude-harness/harness"><img src="https://agentmods.dev/badge/plugins/waqas1412/claude-harness/harness.svg" alt="Measured on agentmods" height="20"></a>
Per session not measured What this adds to a session before it is invoked.
When invoked not measured The manifest loads nothing itself; its 9 skills, 16 agents, 2 hooks cost 2,692 tokens a session between them.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Security

Grade A, and why

harness scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/harness/.claude-plugin/plugin.json · 32 lines

What it actually says

{
  "name": "harness",
  "displayName": "Claude Harness",
  "version": "0.7.0",
  "description": "Portable read-only advisor agents, profile-driven PR/ticket skills, enforcement hooks, and a self-adapting /harness-init that tailors any repo to a documented index pattern.",
  "author": {
    "name": "Waqas Hameed",
    "email": "[email protected]"
  },
  "homepage": "https://github.com/waqas1412/claude-harness",
  "repository": "https://github.com/waqas1412/claude-harness",
  "license": "MIT",
  "keywords": [
    "claude-code",
    "claude",
    "anthropic",
    "agents",
    "subagents",
    "ai-agents",
    "skills",
    "slash-commands",
    "hooks",
    "code-review",
    "developer-tools",
    "productivity",
    "orchestration",
    "scaffolding",
    "conventions",
    "llm"
  ]
}
Contents

What it installs

The manifest is a name and a version. 9 skills, 16 agents, 2 hooks travel with it, and installing the plugin installs all of them — 2,692 tokens a session between them. Each is measured on its own page, and each can be installed alone.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 5d ago First seen · 32 lines scan A 1232f505327b

Subscribe to this mod's changes

harness is a plugin published in the GitHub repository waqas1412/claude-harness (1 stars, last pushed 4d ago), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Building agentsagenticai-agentsanthropicautomationclaudeclaude-code
Related

Other plugins, from other repositories

dx

Developer experience essentials: GitHub Actions debugging, conversation half/quarter-cloning, context handoffs, research (Reddit and Hacker News), private GitHub search, and Claude Code version checks.

ykdojo/claude-code-tips · not measured

junior-flow

Helps junior developers create technical story markdown from a story tracker.

Intai/story-flow · not measured

milestone-driver

Drives a GitHub milestone to completion: triages each issue for design gaps + dependency order up front, then for each buildable issue finds the root cause (or parks it), dispatches a TDD implementer subagent, runs unit + E2E tests + code review, opens a PR, and auto-merges issues on CI green. The whole loop sits…

kenmulford/milestone-driver · not measured

context-optimization

Techniques for extending effective context capacity through compaction, observation masking, KV-cache optimization, and context partitioning. Double or triple effective context without larger models.

viktorbezdek/skillstack · not measured

cc-research

Multi-agent deep research on Claude models: orchestrator -> parallel researcher subagents -> triangulate -> synthesize -> cite -> report. Built-in tools only (WebSearch, WebFetch, Read, Grep, Glob, Write, Task); no external research APIs or MCP servers.

richer-richard/cc-research · not measured

nextjs

Official Next.js skills: adopt and optimize Cache Components, adopt Partial Prefetching, and verify runtime behavior against a running dev server.

vercel/next.js · not measured