agent-prompts

Generate structured prompts for multi-session Claude Code refactoring workflows.

98|3|Updated Jan 22, 2024
One-click install
npx skills add https://github.com/pikax/verter --skill agent-prompts-pikax
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-prompts
Source: https://github.com/pikax/verter/tree/main/.claude/skills/agent-prompts
Command: npx skills add https://github.com/pikax/verter --skill agent-prompts-pikax

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the problem of inconsistent or incomplete agent instructions by providing a standardized, invariant-driven framework for generating prompts that prevent common failure modes in complex refactoring and review tasks.

Core Features & Use Cases

  • Multi-Variant Generation: Supports implementation, continuation, review, and fix-implementer workflows.
  • Stub Prevention: Enforces strict invariants to prevent empty tests, unconditional defaults, and non-discriminating characterization tests.
  • Review Workflow: Automates the creation of a two-session pair (Reviewer + Fix-Implementer) to ensure high-quality, audited code changes.

Quick Start

Ask the agent to generate an implementation prompt for your current refactor plan by providing the plan path and target branch.

Frequently Asked Questions about agent-prompts

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

FAQPage Schema
How do I generate structured prompts for multi-session Claude Code refactoring?

To generate structured prompts for Claude Code refactoring, you provide the repository path, plan file, and branch information to parameterize templates that enforce strict stub prevention and context-handoff protocols for high-fidelity agent execution.

What is the best way to automate code review and fix workflows across agent sessions?

Automating code review workflows is best handled by generating a two-session pair consisting of a Reviewer and a Fix-Implementer, which ensures high-quality, audited code changes while enforcing strict invariants to prevent empty tests and unconditional defaults.

How does stub prevention work in automated agent prompt engineering?

Stub prevention in agent prompt engineering works by enforcing strict invariants within the generated prompts, actively preventing common failure modes like empty tests, unconditional defaults, and non-discriminating characterization tests during refactoring tasks.

Do I need a specific plan file format to drive multi-session agent workflows?

You need a plan file path along with your repository path and target branch information to parameterize the prompt templates, ensuring consistent context-handoff and invariant-compliant execution across implementation and continuation sessions.

Can I generate continuation prompts for an ongoing automated code refactoring task?

You can generate continuation prompts for ongoing code refactoring tasks by providing the existing context and branch details, which creates a structured handoff protocol to seamlessly resume multi-session agent execution without losing fidelity.

Why does my automated agent refactoring produce non-discriminating characterization tests?

Automated agent refactoring produces non-discriminating characterization tests when prompts lack strict invariants; generating invariant-compliant prompts enforces stub prevention to eliminate this failure mode and ensure high-quality code changes.