Skip to content
Back to skills

Revfactory Harness

ASecurity

Claude Code plugin tự động thiết kế agent team + sinh skill từ domain description — 6 architecture patterns, tạo .claude/agents/ + .claude/skills/ tự động. +2098⭐/week.

  • 10 stars
  • 0 votes
  • 0 copies
  • 3 views
  • Added September 9, 2026
ai-agentsbashgit

Works with

  • claude code

Security analysis

A100/100

Scanned September 9, 2026

npx -y skills add yanacuti1121/Yana-AI --skill revfactory--harness --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Revfactory Harness?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Revfactory  Harness
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/yanacuti1121-revfactory-harness/badge)](https://www.skillsdirectory.com/skills/yanacuti1121-revfactory-harness)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: revfactory--harness
description: "Claude Code plugin tự động thiết kế agent team + sinh skill từ domain description — 6 architecture patterns, tạo .claude/agents/ + .claude/skills/ tự động. +2098⭐/week."
allowed-tools: Bash, Read, Write
user-invocable: true
---

Harness (revfactory): factory tạo multi-agent system từ domain description — phân tích domain → thiết kế team → sinh agent definitions + skills tự động.

## Trigger

```
"Build a harness for this project"
"Design an agent team for this domain"
"Set up a harness"
"하네스 구성해줘"  (Korean)
"ハーネスを構成して"  (Japanese)
```

## 6-Phase Workflow

```
1. Domain Analysis        — hiểu requirements
2. Team Architecture      — chọn pattern (xem bên dưới)
3. Agent Definition       — tạo .claude/agents/ files
4. Skill Generation       — tạo .claude/skills/ với Progressive Disclosure
5. Orchestration          — thiết lập inter-agent communication + error handling
6. Validation             — dry-run tests + comparative analysis
```

## 6 Architecture Patterns

```
Pipeline          — sequential dependent tasks (A→B→C)
Fan-out/Fan-in    — parallel independent tasks → consolidate results
Expert Pool       — selective agent invocation theo context
Producer-Reviewer — generation → QA review loop
Supervisor        — central coordinator với dynamic task distribution
Hierarchical      — top-down recursive delegation
```

## Execution Modes

```
Agent Teams (default)
  Dùng: TeamCreate, SendMessage, TaskCreate
  Khi: 2+ agent cần collaborate

Subagents
  Dùng: Agent tool trực tiếp
  Khi: standalone tasks, không cần inter-agent comm
```

## Output Structure

```
.claude/
  agents/
    <agent-name>.md    ← agent definitions
  skills/
    <skill-name>.md    ← generated skills với Progressive Disclosure
```

## So với yamtam harness

Harness này: generate từ domain description (top-down)
yamtam harness: pre-built curated patterns (bottom-up library)

## Source

https://github.com/revfactory/harness · MIT · +2098⭐/week

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.