skills.rest
enkira-ai
enkira-aiOFFICIAL
Explore all skills by this author →
WebsiteGitHub

harness-init

Analyze a repository and generate CLAUDE.md, features.json, progress.txt, init.sh, and docs/ scaffolding.

2|Updated Mar 7, 2026
Search Tags:#workflow#scaffolding#claude#agent#repository#autonomous#harness
One-click install
npx skills add https://github.com/enkira-ai/claude-plugins --skill harness-init-enkira-ai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: harness-init
Source: https://github.com/enkira-ai/claude-plugins/tree/main/plugins/harness-engineer/skills/harness-init
Command: npx skills add https://github.com/enkira-ai/claude-plugins --skill harness-init-enkira-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

You are the initializer agent. Your job is to analyze this repository and create the scaffolding that enables coding agents to work autonomously across many sessions without human supervision.

The harness solves problems of context loss, one-shot work, and architectural drift by generating a legible, structured environment with clearly defined outputs that guide each session.

Core Features & Use Cases

  • Analyze the repository to determine language, tests, and existing structure.
  • Create the full harness scaffolding (CLAUDE.md, features.json, progress.txt, init.sh, docs/) so agents can operate independently across sessions.
  • Provide a repeatable setup workflow that can be reused across multiple repositories and sessions.

Quick Start

Invoke the harness-init skill on a repository to generate the full harness scaffolding (CLAUDE.md, features.json, progress.txt, init.sh, docs/) and prepare autonomous agent sessions.

Frequently Asked Questions about harness-init

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I set up a repository for autonomous coding agents to work without human supervision?▼

Autonomous agent scaffolding analyzes your repository structure and generates standardized entry points and session workflows, including CLAUDE.md, features.json, and init.sh, to prevent context loss and architectural drift across sessions.

What is autonomous agent scaffolding and how does it prevent architectural drift?▼

Autonomous agent scaffolding analyzes your repository structure and generates standardized entry points and session workflows, including CLAUDE.md, features.json, and init.sh, to prevent context loss and architectural drift across sessions.

What files do I need to scaffold a repo for Claude agent sessions?▼

Scaffolding a repo for Claude agent sessions generates CLAUDE.md, features.json, progress.txt, init.sh, and a docs/ directory to provide a structured, reproducible development environment for unsupervised work.

Does this scaffolding workflow require any specific dependencies or components?▼

This scaffolding workflow has no external dependencies or components; it analyzes your existing repository language and structure to directly output the required harness files for autonomous agent operation.

Can I reuse the same scaffolding setup across multiple different repositories?▼

Yes, the scaffolding setup provides a repeatable workflow that standardizes the autonomous agent environment and outputs across multiple repositories, ensuring consistent session management regardless of the specific codebase.

Related Skills

Explore More Software Engineering Skills→
sd0xdevsd0xdev
187

post-dev-test

Analyze developed features for integration and end-to-end test coverage gaps.

Community
Intermediate
KarlinskySKarlinskyS

opencode

Automate coding tasks with the OpenCode CLI for implementation, refactoring, and PR reviews.

Community
Advanced
HouseGarofaloHouseGarofalo
3

autonomous-agent-harness

Automate autonomous coding agent project setup with Archon MCP state management.

Official
Advanced
jrodrigopucajrodrigopuca

analyze-project-scope

Analyze repository structure to define documentation scope and complexity.

Community
Intermediate