wrought-implement

Coordinate iterative code changes with automated verifier runs and loop-state tracking.

3|2|Updated Apr 3, 2026
One-click install
npx skills add https://github.com/fluxforgeai/ARTEMIS --skill wrought-implement
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: wrought-implement
Source: https://github.com/fluxforgeai/ARTEMIS/tree/main/.claude/skills/wrought-implement
Command: npx skills add https://github.com/fluxforgeai/ARTEMIS --skill wrought-implement

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automates iterative implementation cycles by coordinating code changes, running a verifier, and maintaining loop state so developers can converge on passing tests without manual bookkeeping or repeated context switching.

Core Features & Use Cases

  • Active-loop detection: Prevents overlapping runs by reading .claude/wrought-loop-state.json and halting if an active loop exists.
  • Plan discovery and confirmation: Locates the most recent docs/plans/*.md plan when a plan path is not supplied and prompts for confirmation before proceeding.
  • Verifier-driven iterations: Reads verifier configuration from the .wrought marker (verifier command, test path, max iterations, timeouts), initializes a loop state file, executes the verifier after each implementation attempt, and enforces iteration budgets and timeouts.
  • Stateful automation and reporting: Writes detailed loop metadata to .claude/wrought-loop-state.json, captures iteration history, and summarizes completion with artifacts placed under docs/capsules/{finding_id}/.

Quick Start

Start an autonomous implementation loop using the most recent plan in docs/plans and follow prompts to confirm verifier settings.

Frequently Asked Questions about wrought-implement

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

FAQPage Schema
How do I automate test-driven implementation loops for repository code changes?

Autonomous implementation loops coordinate code changes and automated test verification by reading verifier commands and timeouts from a .wrought marker, executing iterations until tests pass or max limits are reached.

What do I need to set up before starting an autonomous test loop in my repo?

Before starting, create a plan file under docs/plans and a .wrought marker defining verifier commands, timeouts, and iteration limits. The loop checks .claude/wrought-loop-state.json to prevent overlapping active runs.

How does active loop detection prevent overlapping autonomous implementation runs?

Active loop detection reads .claude/wrought-loop-state.json and halts initialization if an active loop exists, preventing overlapping autonomous implementation cycles from conflicting during code changes and test verification.

How are verifier commands and iteration limits configured for repo automation?

Configure verifier commands, test paths, max iterations, and timeouts in a .wrought marker file. The autonomous loop reads this marker to execute the verifier and enforce iteration budgets after each implementation attempt.

What happens when max iterations or timeouts are reached during test-driven loops?

When max iterations or timeouts are reached, the loop enforces its budget and halts further attempts. Iteration history and loop metadata are captured in .claude/wrought-loop-state.json, and completion artifacts are placed under docs/capsules.