rb-create-low-level-plan

Compile approved implementation phases into schema-3 plans with static file-state verification.

Updated Jul 2, 2026
One-click install
npx skills add https://github.com/richardmbailey/rb-skills --skill rb-create-low-level-plan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rb-create-low-level-plan
Source: https://github.com/richardmbailey/rb-skills/tree/main/rb-create-low-level-plan
Command: npx skills add https://github.com/richardmbailey/rb-skills --skill rb-create-low-level-plan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pydantic, pydantic-ai-slim, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This skill addresses the need for high-integrity, statically verifiable implementation plans for constrained development phases, preventing unauthorized execution or unsafe operational drift.

Core Features & Use Cases

  • Static Verification: Ensures all success criteria are observable through file-state inspection, rejecting executable or runtime-dependent checks.
  • Bounded Operations: Generates proposal-only tasks that return strict unified diffs, ensuring the model never receives direct write or patch-application authority.
  • Use Case: When working on a sensitive codebase, use this skill to compile a specific, approved implementation phase into a schema-3 plan that guarantees safety through static analysis and explicit resource grants.

Quick Start

Invoke the rb-create-low-level-plan skill to compile the current approved implementation phase into a typed operational contract.

Frequently Asked Questions about rb-create-low-level-plan

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

FAQPage Schema
How do I create statically verifiable implementation plans for constrained development environments?

Statically verifiable implementation plans are created by compiling approved phases into schema-3 proposals that enforce bounded agent tasks, returning unified diffs while rejecting runtime-dependent checks to ensure operational safety and auditability.

What is static file-state verification for agent task planning?

Static file-state verification ensures all success criteria are observable through file inspection, rejecting executable or runtime-dependent checks to maintain high-integrity, auditable implementation plans without unauthorized execution.

How do I prevent unsafe operational drift when generating agent implementation tasks?

Unsafe operational drift is prevented by generating proposal-only tasks that return strict unified diffs, ensuring the model never receives direct write or patch-application authority over the codebase.

Can I use pydantic to enforce explicit resource and provider grants in implementation plans?

Yes, pydantic is used to enforce explicit resource and provider grants by compiling approved implementation phases into typed operational contracts that bind agent tasks to strict, statically verifiable proposals.

What are the limitations of proposal-only unified diffs in constrained development?

Proposal-only unified diffs limit the model to generating strict patches without direct write or patch-application authority, requiring external systems to apply changes and preventing autonomous execution in sensitive codebases.